Most Roadways Aren't Multi-Modal

After someone can be made to recognize that a corridor where cars occupy 99% of the space and have (effective) priority at all conflict points, and admits that cars are being prioritized, they will then attempt to weasel their way out. After several years arguing with goons on Twitter, this usually consists of series of disputable claims:

All Roads are Multi-modal!

Dream on. If it was actually multi-modal, it would have a dedicated lane for that mode, and design standards that reflected all modes, and managed conflict between them.  If I design a home for horses, no one tries to pretend it's for humans, even when humans can use it. There are no 45mph humans, so no road with a 45mph design standard is designed for humans.

Sidewalk, Curb, Gutter, the park-strip and 'Safety Area' are Multi-modal!

Bunkum. Curb and gutter exist to funnel water aware from the road, to ensure that water flowing off the road doesn't undermine the roadbase* that supports the asphalt. Even if you didn't need to have a sidewalk, a road would still need to build it. And the 'Safety Area' between the right-most travel lane and the curb and gutter, which gets used for bike lanes? It's for car safety. Older highway manuals will still give it the proper name, which is "recovery space", and "obstacle free zone" and it was intended to give an out-of-control car time to recover before it ran off the roadway and crashed. In urban areas, it's been coopted for on-street parking, resulting in lots of crashes. The 'park strip' is also a misnomer, because it's actually where obstacles get put--telephone poles, traffic cabinets, bus stops, and sometimes some grass and small, bedraggled trees**. These things would be fine to have in the 'safety area', except cars. So, for your standard arterial, that's four lanes of auto traffic, one turn lane, two 'recovery space' areas, curb and gutter, the park strip, and two 5' sidewalks. Which means 95% of space is getting devoted to cars. 

Curbs exist to Protect Pedestrians!

Nonsense. I drive an SUV, and I drive over the curb every time I try to parallel park. For an out-of-control car, a curb is barely a speedbump. Bollards, now those protect pedestrians. I can offer no better evidence than 7-11, which clearly understand that stopping a moving car requires a massive metal pole reinforced with concrete, embedded four feet into the ground. 

*roadbase: specialized compacted soil that can hold up heavy weights. Often a mix of crushed rock, gravel, pea gravel, sand, and maybe some clay. Largely free of organic matter (which is why street trees die). 

**Not by accident. Trees get planted in roadbase, which is basically free of any kind of nutrient, so they don't grow. Which is fine for highway enginneers anyway, as they  don't want roots disrupting the road base. (Many state codes require cities to put trees in concrete boxes to constrain roots. And as trees are as large above as they are below, a 5'x'5 concrete boxes creates a 5'x5' canopy above. 

Modal balance & multi-modal corridors

Modal balance is a dead concept, a weird rhetorical legacy of past planning theory. Planning now is about which modes do we prioritize and where. I explicitly differentiate between the old paradigm of 'share the road', where all users are expected to make use of arterial roadways, and what I'll call the "Streetmix" paradigm (after the website), which says "we have 60' of ROW, how are we allocating that to car lanes, bike lanes, landscaping, sidewalks, and street trees?", and realizing that there is often more value created by not following the historic default of prioritizing auto capacity uber alles.

More broadly, it gets into the issue of what it means to have a multi-modal corridor. It's sometimes used in minimalist way, to suggest where it's possible for multiple modes to be present. Most corridors are nominally multi-modal, in the sense that it's nominally legal for other modes (pedestrians, bikes, etc) to travel along them (freeways being the exception). But at a higher standard, having a multi-modal corridor means having infrastructure for multiple modes -- sidewalks for urban streets, or a divided highway that has bus stops. 

But for anyone who actually makes use of non-auto infrastructure, non-auto modes are clearly second class. Because for transportation, it's the network that matters. Sidewalks that end mid-block aren't very useful. Bus stops where the ADA-compliant sidewalk stops beyond the bus stop are a bit of a cruel joke.  So the next tier of quality for a multi-modal corridor is one that is designed to encourage use by multiple modes, rather than merely permit, tolerate, or minimally accommodate them. 

Recognizing that automobile access has been prioritized over that of other modes is the first step, because it opens the door to the option that other modes might be similarly prioritized. Traffic planners intuit that is dangerous, and so there will be endless whatsaboutism that the corridors are already multi-modal, designed to obscure the fact that most planning prioritizes the automobile. 

After someone can be made to recognize that a corridor where cars occupy 99% of the space, and have (effective) priority at all conflict points, and admits that cars are being prioritized, they will then attempt to justify that priority. After several years arguing with people making goon arguments on Twitter, these will include the chestnut that the gas tax pays for it. If they can be made to accept what the gas tax actually pays for (ie, not the existing ROW), an awkward silence may ensue, and then it's possible to have a real conversation about how we allocate an expensive, limited public resource to maximize transportation and economic development benefits, and what that means about prioritizing different transportation modes. 

Honolulu Rail

The most interesting bit about the Honolulu train was it's proposed automation: no driver, just a program. With elevated guide-way, there is nothing to crash into on the track, so all the driver does is stop and start the train--the rail guides it. No reason not to automate that. But it's what automation makes possible that is interesting: increasing frequency is no longer dependent on labor costs, but merely a matter of capital development. Running more trains per hour simply requires buying more trains. Theoretically, this could make very high frequency trains (ever 2-3 minutes) technically possible. And thus enable a transit-oriented lifestyle difficult (if not impossible) to duplicate elsewhere.

It would cost money. But it could radically reshape urbanism in Honolulu. Moreso than the area around the line itself, because a high-frequency transit spine enables the emergency of a transfer network, making is possible to get between places that aren't on the rail line, but are connected to it by bus/shuttle routes.