Today’s Headlines

  • Jeff Klein and Co. Set to Screw New Yorkers on Congestion and Transit (PoliticoNews)
  • Other Cities Are Looking to New York to Lead on Road Pricing (WSJ)
  • Justin Davidson Pulls No Punches in Pricing Q&A (NY Mag)
  • Contractors Say They’re Not to Blame for Exorbitant MTA Construction Costs (Crain’s)
  • Byford Will Take Questions on Twitter Tomorrow (Post)
  • Bronx Residents Buck Cuomo’s Plan to Botch the Sheridan Teardown (City Limits)
  • AMNY, NY1 Cover Program to Allow Cyclists to Proceed on LPIs
  • MTA Worker Hits and Kills Colleague With Bus at Queens Depot (News)
  • Motorist Critically Injures Woman on Linden Blvd. in Brownsville; Cops, News Blame Victim
  • News: “Out-of-Control Car” Takes Down Scaffolding on Second Avenue in Harlem

More headlines at Streetsblog USA

  • J

    If you read the Sheridan headline too quickly, it sounds like “Bronx residents BACK Cuomo’s plan”. I was upset and confused, but luckily that’s not the case. 🙂

  • Tooscrapps

    Don’t worry, the upstate casinos that aren’t making enough might get some aid. Perhaps Cuomo can shift some MTA money there, where it’s needed most.

  • Larry Littlefield

    So how much of Nassau and Suffolk’s sales tax revenue will be diverted? And how much of New York City’s sales tax revenue will be used for the LIRR?

  • Fool

    lol. What is NYC going to do about it? Have competitive elections? Competitive primaries?

  • Larry Littlefield

    This is just the start. Once those pigs in the legislature get started on something, such as pension increases or tax breaks for developers, it rolls on into “me too.”

    How about a lawsuit? Not just about this, but about everything.

  • bolwerk

    Short of something akin to proportional representation, competitive elections aren’t very helpful. The GOP is even more dogmatically neoliberal than the Democrats, and would probably only serve to pull the Dems even further to the right.

  • Larry Littlefield

    Hey Bolwerk, take off you ideological binders and realize that has nothing to do with it.

    Since when is subsidizing upstate casinos with NYC sales tax money right wing?

    It’s people like us trying to get over on people like them. Purported ideologies are extremely flexible to the point of meaninglessness.

  • AMH

    The AMNY headline on the web (“NYC bike safety program allows cyclists to follow pedestrian signals at 50 intersections”) is quite a far cry from the sensationalist ones in print (“Bikers can now legally run some reds” and “No stopping NYC bikers”). I was thankful that the actual story was more nuanced and actually quite accurate.

  • bolwerk

    It literally doesn’t get much more right-wing than stealing from poor people in NYC to provide welfare to casino moguls upstate.

  • Vooch

    Powerful in-depth article on Uber self driving in City Lab today by Laura Bliss

  • Vooch

    the left-right paradigm is completely faux designed to obscure rather than illuminate

  • bolwerk

    It makes perfect sense if you know what the terminology actually means. I prefer to avoid it, but I didn’t see another way of saying what I said concisely.

  • bolwerk

    All this masturbation about self-driving cars, and there is almost no discussion of the social implications. If they’re going to be deployed on the grand scale, how do you do it?

  • Vooch


    what social implications ?

  • Fool

    “‘Out-of-Control Car’ takes down scaffolding” seems like another great reason to get rid of this city’s ridiculously corrupt scaffolding law.

  • Larry Littlefield

    “Both former operators also expressed surprise that Uber’s self-driving technology had failed, too. Something went deeply wrong here, they said—the cars were often super-sensitive to obstructions, even non-existent ones. “Sometimes the car would brake because of steam coming up from a pothole in the ground,” the anonymous operator said.”

    “Self-driving experts are equally perplexed. “The LiDAR should have seen the pedestrian at quite some distance,” Raj Rajkumar, who leads autonomous vehicle research at Carnegie Mellon University, wrote in an email. “All in all, this indicates pretty serious technical issues at the core of the Uber system.” Velodyne, the company that designed the LiDAR sensors used by Uber’s self-driving fleet, told the BBC it was “baffled” by the crash.”

    I grind on a PC all day long. On occasion, the machine just sort of slows down and locks up. IT always has the same solution, reboot, which always works.

    I wonder if the computer had sort of stalled at the moment of impact, like a human driver zoning out or taking their eyes off the road. That would explain it. The system is not “fail safe.”

  • bolwerk

    Nine figures of drivers who aren’t going to want their steering wheels dragged from their cold, dead hands?

  • reasonableexplanation

    You don’t think society will fundamentally change (for better or worse) once most cars are self driven?

    Consider how society changed when we went from railroad to car as the main form of transport.

  • Vooch

    level 4 accommodates a mix

    plus most car manufacturers notably Ford and Toyota already are providing ‘neural node’ capability in their human driven cars as standard equipment.

    My thinking is AVs will have little social effect other than making 1 car suburban households more likely.

    You should watch the Ford CES most recent keynote

  • vnm

    At first, if they don’t want a self-driving car they don’t have to buy one. Later, they’ll see the enormous insurance break they’ll get by shifting, and their concerns may become less passionately felt.

  • bolwerk

    Self-driving cars aren’t going to be particularly useful until everyone is using one. Somehow that transition needs to be managed.

  • bolwerk

    I didn’t say the social implications are bad, but they’re going to be huge. And you can probably look to NYC congrestion pricing to see what the resistance will look like.

  • vnm

    Ehhh. I don’t know. I think it would be an organic evolution that defies attempts at top-down management. So far the technology has led the regulators, not vice versa, and I think that will continue. But the fundamental incentives will be there to continuously push it forward to grow market share, then reach a tipping point where it becomes the norm.

  • Vooch

    Human driven and AVs can mix on same street.

    it’s not an issue

  • bolwerk

    So far the technology hasn’t particularly impacted anything either. That will need to change.

  • bolwerk

    At the very least it’s a huge issue if you want one of the most useful benefits of automation, the capacity-increasing kind like vehicles driving closer together.

  • It is an issue, because the self-driving cars will be following the law and will be braking appropriately (the recent Uber disaster notwithstanding), while the human drivers will continue to behave like socipaths. So those humans will have an advantage. For example, a self-driving car that tries to enter a highway will have a hard time if the cars already on the highway are all driven by humans.

  • More likely than an inconveniently-timed software glitch on the part of the lidar sensors is some engineering failure on the part of Uber.

  • Larry Littlefield

    It could be an inconveniently timed glitch in the Uber computer. The data was coming in, but the computer briefly stopped controlling the car, so it kept going.

  • crazytrainmatt

    Reliability of the level we would consider safe requires a dedicated safety culture and a very different engineering approach. The hallmark examples of complex technical operations with excellent safety records — commercial aviation, nuclear power generation, and naval submarines — look nothing like consumer electronics or the present development of self-driving cars.

    These fields have a deep appreciation for human factors (look at the screenshot of the Hawaiian system that sent out the false incoming nuke warning for an example of how not to do it). None of these fields would ever countenance a paradigm that relies on a mostly disengaged human driver to instantly reacquire situational awareness and react. Read the NTSB report on the AF447 crash for an example of how this lesson was re-learned in the aviation community.

    The 90/10 rule in engineering says that the first 90% of the functionality takes only 10% of the effort. So cars can drive themselves in ideal conditions but the overall system is brittle and fails upon slight deviations from the training set. Silicon Valley has a long tradition of vaporware (overpromises and carefully orchestrated demos, but the shipped product only materializes dramatically later).

    It seems that Uber is the most reckless of the bunch, but all of the self-driving car implementations will run into this. The risk here is (as you’ve said), that the safety of those outside the cars is bartered away, or that cities are rebuilt to fit within the limitations of self-driving cars like when the interstates were built.

  • Vooch

    true but a counter argument is AVs will make it difficult for human cabbies to drive 50 on 1st avenue

  • Larry Littlefield

    Tesla’s stock and bonds have crashed after a possible self driving crash killed somebody inside the vehicle.

    I don’t recall Uber’s stock falling as much after its vehicle killed someone outside the car. Given the woman was poor, the company might file a claim against the State of Arizona for a share of the savings on her future health care costs.

  • Good point.

    I am of the opinion that automated vehicles will do more good than harm. Still, I am quite sure that there will be some challenges during the long transition period that we haven’t even conceived of yet.

    In addition, we must acknowlege the possible negative knock-on effect: namely, an overall increase in auto use, which in turn could lead to still more street space being devoted to cars at the expense of other uses. And then there’s the question of promoting sprawl. These are matters that will have to be handled by means of regulation.

  • bolwerk

    I agree.

    There is the possibility of more being done with fewer vehicles, which would be good. But that’s not what automakers are going to want, and anybody who says special interests are organically going to let the best possible future happen hasn’t been paying attention to the past two centuries of transportation politics.

    I figure reduced need for parking is going to be a boon for urbanists.

    But a possible drawback, if things aren’t handled intelligently, is more vehicle-miles traveled and more energy consumption – now with deadheading cars!

  • crazytrainmatt

    Uber is a private company, so their valuation is typically inferred from the last funding round and doesn’t reflect recent events. They lost $5B last year dumping capacity and this is not their first black mark. It will be a miracle if they can turn that company around.

    But Tesla’s system is almost riskier for the reasons I mentioned above: automation must be 100% reliable since a human driver can not be expected to reestablish control in time to act safely. ABS brakes meet this standard (they basically always work), whereas lane-following systems can be enabled in circumstances that can easily deteriorate past their design limits.

    It would be a huge miss if Uber and Tesla take sole blame for these incidents. Instead, this should be a wakeup call to slow down blanket enabling legislation until it includes reasonable technical and reporting requirements. Note the poor quality video initially shaped the Uber narrative as “no human driver could have avoided her”, when in reality anyone watching the road would have braked and avoided her.

  • AnoNYC

    Got this email:

    Dear Via-istas,
    Leaders in Albany are proposing a dramatic new tax on Via trips in Manhattan, on top of the existing sales tax. This will affect the hundreds of thousands of New Yorkers who depend on Via as a reliable, affordable way to get around the city.

    If our Albany representatives are serious about fixing congestion, they wouldn’t penalize those who share rides. Any plan to tackle congestion in New York should not involve raising taxes on shared services.

    We need your help today. Make your voice heard by signing our petition to state legislators.

    Sign Here
    With love,

    Team Via NYC

  • AnoNYC

    Cynthia Nixon endorses ‘comprehensive’ congestion pricing to ease transit funding crunch

  • Larry Littlefield

    One reasonable approach would be to allow full autonomous control only on limited access highways and, perhaps, rural highways between towns.

    With the system as a back-up for human drivers in places with more pedestrians and cyclists, to avoid the horror we saw in Park Slope.

  • Larry Littlefield

    Makes sense except Manhattan is a place where more passengers per vehicle than Via provides are justified.

    Wipe out the bus system in the outer boroughs if you want, given what the TWU and state legislature have arranged for buses to cost. Try to replace the subway and express buses to Manhattan and you’ll see it is mathematically impossible.

  • reasonableexplanation

    I don’t follow your reasoning… It’s trivial to enter a highway, even with the short ramps that NYC has, as long as you use it as intended and actually accelerate, which AVs should be able to do. The only time there are issues is when a timid or inexperienced human driver doesn’t do that, and ends up stopped at the end of one.

  • Once a highway is packed such that there are no gaps between cars, entering it requires that someone already on the highway let you in. A human driver can encourage this by nosing ahead; an automated driver would be at the mercy of human drivers, and could be stuck sitting on an on-ramp for a considerable length of time.

    In reports that I have read and seen of Uber’s experiments with automated vehicles on the streets of Pittsburgh, the question of left turns has been mentioned. The automated vehicles have to contend with drivers’ practice of making a left turn as soon as the light changes, which they do instead of waiting until the cars coming in the other direction have passed. (And some people want to tell me that drivers are not sociopaths!) The automated vehicles are at a disadvantage in this case, as they are in every case in which their law-abiding practices meet up with drivers’ tendencies to cheat.

    One can only hope that the advent of self-driving cars will lead to a renewed commitment to traffic enforcement, so that the automated vehicles will not be systematically stymied by drivers’ bad behaviour.

  • Vooch


    Your points are well taken and I agree that AVs are not a solution to the fundamental problem of private cars simply consuming vast amounts of space versus other modes.

    However, my sense is the momentum of reallocation street space towards more efficient uses will continue AVs or no AVs.

    The economic costs of subsidizing private cars is simply too great for cities.

  • Joe R.

    Everything you wrote is exactly why once autonomous vehicle technology becomes a little more mature we must rapidly disallow human driving on public roads. AVs will have myriad benefits, such as freedom from traffic jams, near zero collisions, higher speeds on highways, no need for traffic controls, etc. BUT none of these benefits can be realized so long as we have a mix of human and AI driven vehicles on the roads. The so-called “transition period” needs to be limited to a handful of years. I’d say by 2020 all of the major kinks in AVs will have been worked out. After that, maybe you pass a law stating all new vehicles sold won’t have manual controls. Existing vehicles will have to either have their manual controls removed by 2025, or they will no longer be allowed on public roads. After 2025 anyone caught driving manually on public roads would have their vehicle confiscated.