21 ms·

The War on Waze

134 points by Fins 9y ago

PeterStuer 9y ago
Driving on neighborhood streets is asocial and callous. Wether you use an "app' or not is irrelevant, you're still wrong. As social community feelings have eroded, I guess we now must use technology to combat those that are not polite enough to comply. Make all but the roads designed for transit prohibited to all but local traffic, enforce by cutting streets or ANPR cameras.
- seanmcdirmid 9y ago
  It’s all good if your roads are privately funded. If they are funded by taxpayers though...
  - ghaff 9y ago
    Not everything is solely about legality.
  - maxerickson 9y ago
    If they are funded by taxpayers taxpayers can set rules for their use?
    - seanmcdirmid 9y ago
      Yes, as long as the minority doesn’t rule the majority.
- c22 9y ago
  While I agree with your sentiment, technologically barring residential streets for non-local traffic would have a lot of downsides anytime your friends want to visit or you order a pizza.
  - zachkatz 9y ago
    That counts as local traffic
    - c22 9y ago
      Yes, that's my point.
- rootlocus 9y ago
  > Wether you use an "app' or not is irrelevant, you're still wrong. So I'm wrong to drive on public streets built with public funds because you feel entitled to it by living on close proximity? It might be inconvenient for you, but it's not wrong. > Make all but the roads designed for transit prohibited to all but local traffic As long as the prohibited roads are payed for by the locals, sure. There are plenty of residential complexes with private roads, but don't pretend like public areas are private.
ghaff 9y ago
Waze is actually only one of the topics the article covers. But it’s hard to see what anyone can do. It’s arguably obnoxious to routinely route people on quiet residential streets to save a few minutes. However, they are public streets and experienced commuters might well do the same routing.
- GCU-Empiricist 9y ago
  So does Waze do something more aggressive than Gooogle's faster route option. This is the second article Ive seen blaming Waze, but I get similar recommendation's as a casual Google Maps user who checks which of three routes is fastest in each morning.
  - addicted 9y ago
    I always assumed that Google used Waze data/algorithm for their maps? I would be surprised if they bought Waze, but didn't integrate it with Maps at all.
    - Bedon292 9y ago
      Waze is definitely more aggressive about its neighborhood routing. Though they could be using the same underlying code somewhere, its definitely tuned differently. The main integration point into Google Maps is the accident reporting and active traffic data.
      - sjwright 9y ago
        In my experience, I find that Waze seems to underestimate the time cost of intersections. I think that's why it is more willing to route you through local streets than its competitors.
        TheAdamist 9y ago
        I find it overestimates traffic lights sometimes, on a medium road next to a grid neighborhood to turn right on another medium road, generally has you cut the corner by one block to avoid the light, even if you can't ever make it out into traffic before the light changes anyway. Try and safely pay attention to wazes route and ignore it sometimes, but it's generally correct. And you can send audio prompts to a phone speaker instead of Bluetooth, unlike gmaps, so that's also useful.
enoch_r 9y ago
As someone who leans quite strongly libertarian, I do not understand what seems to be the default libertarian position on roads. The default view seems to be that government-funded highways are essentially a god-given right. If I want to travel 30 miles every day from home to work, well, somebody had better build me a 8-lane highway to get me there with a minimum of congestion. (Sure, maybe sometimes we'll implement congestion pricing that will cover a miniscules portion of the cost of building and maintaining the highway. This is still a massive government subsidy for people who drive.) Bike lanes, on the other hand? Public transit? Those get the more typical libertarian viewpoint on government spending. How dare the government try to distort our incentives by spending millions of dollars to provide bike infrastructure?! (Fun fact: As of 2008, the replacement cost for all of Portland's bike infrastructure was estimated at $60M, the rough cost of one mile of urban freeway. I'm sure it's gone up by then--it's still a rounding error.) It's genuinely weird how libertarians--generally not ones to ignore the effects of government spending on incentives--don't appear to notice that spending hundreds of billions of dollars on a transportation system that you can only use if you're in a car might actually, you know, cause more people to drive.
- Analemma_ 9y ago
  > The default view seems to be that government-funded highways are essentially a god-given right. What sort of libertarians are you hanging out with? I'm not a libertarian myself, but most of the ones I have talked to about roads wouldn't agree with this. Their usual position is that roads should be privatized and tolled, and that this would help fix the externalities (traffic, concrete deserts, suburban sprawl) caused by "free" access to roads, if people had to pay their true cost.
  - acdha 9y ago
    I think this is an example of real versus expressed preferences: I’ve known a bunch of libertarians who would when thinking about the issue say all roads should be toll roads but at all other times act as if they should be taxpayer funded, complain that their taxes are way too high and should pay for plenty of roads, talk like construction prices shouldn’t have gone up since 1960, etc. Their unconscious way of thinking about it was basically the same as everyone else, and I think that’s in part because like everyone else they didn’t really want to pay every time they drive somewhere.
hprotagonist 9y ago
I remember reading some years ago about a village in the UK that wanted to be removed from satnav maps. At that time, most of the people who used them were truck drivers, and the 12th-century village's main road was about 3 inches too narrow to allow a truck through. A wedged truck on main street a week for 8 months, and people got pretty cranky about it. I'm not sure what ever happened, though.
- reificator 9y ago
  > A wedged truck on main street a week for 8 months, and people got pretty cranky about it. I think you accidentally a word there.
  - ocdtrekkie 9y ago
    Not sure it's actually wrong, but it does read weird. "A wedged truck on main street [each] week for eight months" reads a little better, but I don't think what the parent comment said was technically incorrectly written.
    - hprotagonist 9y ago
      Perhaps I should’ve said “One truck” instead of “A truck”.
      - ocdtrekkie 9y ago
        That would've read a little easier as well.
        hprotagonist 9y ago
        the fancy-pants term for this, i think, is https://en.wikipedia.org/wiki/Garden_path_sentence https://en.wikipedia.org/wiki/Garden_path_sentence
  - lainga 9y ago
    It's a clipped version of After putting up with [one] wedged truck..., it seems valid to me. Similar to An apple a day keeps the doctor away, or more similarly, An apple a day, and the doctor stays away.
ndaiger 9y ago
I live in Los Angeles near a lot of the construction going on for the new subway line running down Wilshire Blvd. My neighborhood (Carthay) has been getting a lot more cut-through traffic. There are days that seem particularly "Waze-y", when it's basically just gridlock on our little streets for an hour. I imagine it's due to apps re-routing people. But I don't really think it's any faster for the commuter. I'm reminded of this article (seemingly now password-protected, so I'll link to a summary) that Waze chronically underestimates your travel time, while Apple Maps chronically overestimates (and therefore overdelivers): https://appleinsider.com/articles/18/02/22/informal-testing-finds-apple-maps-arrival-times-are-intentionally-conservative-to-provide-a-good-user-experience https://appleinsider.com/articles/18/02/22/informal-testing-... From my anecdotal experience, Waze does vastly overestimate the benefits of a crazy route, like making a left turn from a stop sign across 6 lanes of rush-hour traffic.
- ghaff 9y ago
  Google Maps does this too at times. It isn’t so much residential streets but I’ve had Google take me on windy country drives where I inevitably screw up.
- wyclif 9y ago
  It's ludicrous that Los Angeles politicians are blaming Waze for this, when it's their own fault for failing to approve and build higher-volume and better roads. I smirk every time I read some bureaucrat blaming apps that demonstrably make our lives better instead of owning up to their own impotence and incompetence.
  - bsagdiyev 9y ago
    Have you been to Los Angeles? On what space do you expect them to do this on, without interrupting the already bad traffic and with what money? I'm genuinely curious as to your idea and how it would work.
    - wpietri 9y ago
      And even if they somehow do manage to do it, I'm not sure at all that any lasting improvement will be made. An urban planner friend tells me that people decide where to live based on commute time, not miles driven. So if you put in more roadway, people move to cheaper places farther out, using more road-miles than before, at least until the speeds drop back down.
jcranmer 9y ago
For city planners, there's a natural hierarchy of roads: limited-access freeways, arterials, collectors, and residential streets. Moving high volumes of cars is intended to use arterials and freeways, whereas collectors and residential streets are deliberately designed for lower speeds as they may have a higher likelihood of pedestrians randomly entering the street (the child racing after his ball, to use the proverbial example). In a nutshell, the problem with Waze is that it encourages drivers to shift from arterials to collector roads, where they very often proceed to drive in flagrant violation of safety standards. Getting drivers to not do that is challenging (and this problem predates Waze). Speed bumps don't meaningfully slow people down. Even prohibiting the through movement in a T intersection by means of an island doesn't work--people will happily drive over the curb to do so. By routing people down roads that are not designed for this traffic, Waze is pretty close to being complicit in these safety violations. The proximate cause of this furor is that there is a street in LA that is too steep to be safe for through traffic, yet Waze is happily routing people along the street. Waze's response is, effectively, it's legal for through traffic, so Waze doesn't have to do anything, with the underlying reasoning being that it doesn't want to have to deal with everyone demanding special treatment. Waze's attitude is, quite frankly, irresponsible. What they could do--and should be doing--is to look at potential factors that impact safety and update they app to not route people on roads that are clearly unsafe (Google's maps already have the ability to report grade on roads, for example). In fact, that's what the LA council is basically doing: they're looking to see if Waze could be doing more, and in the likely case they find that Waze is indeed not doing enough, finding some mechanism to compel Waze to do more. The other element in play here is a complete about congestion. The truth is... you can't fix congestion. In transportation theory, there's an element called induced demand: if travel time improves in a corridor, then that induces people to take extra trips, often returning it fairly quickly to the same level of congestion. What many cities have realized is that trying to provide sufficient road capacity to meet demand is a game lost before they've even begun, and instead focused their efforts on trying to coax better utilization of current corridors (single-occupant vehicles on roads are basically the lowest-throughput transportation measures devised, except for Elon Musk's proposals which somehow manage to be even worse).
- galdosdi 9y ago
CarVac 9y ago
I live in Leonia, as mentioned in the article, and while I'm unsure of how much less traffic businesses are actually seeing, I do notice that when I'm navigating to places out of town, Google Maps and Waze never route me on local streets inside town anymore, preferring to go on the main thoroughfares until outside the town, and then crossing between them on other towns' local streets.
dkural 9y ago
It's ironic that the article seems to think building more roads, thereby improving road-to-miles-driven ratio will ease congestion. It also takes as fact that it's "as inefficient" to use Waze. My experience is Waze often does cut commute time down; not always; but on the aggregate it demonstrably, scientifically, provably does. That's why it's a huge success and people use it. I do think there are problems with drivers disregarding the law (speed limits, signs etc.) on smaller streets, and perhaps they need to be more aggressively punished for doing so. In general I am for more aggressive fees, sentencing and punishment for violating driving rules.
- steve_musk 9y ago
  Where is the scientific proof? Legitimately curious.
- Fins 9y ago
  But did you have somebody leaving with you, and NOT using Waze to prove that Waze does cut commute down compared to something else? Disclaimer: I usually do use Waze commuting, but find its suggestions in certain places downright bizarre.
  - Casseres 9y ago
    Waze is now owned by Google, but I don't know if they share the same route-suggestion engine. I've had Google Maps (desktop web browser) give me a route once through heavy traffic, but when I dragged the route to a mostly-parallel highway, my estimated time decreased by 10 minutes. So it's not perfect, or they wanted more users to give more data on the traffic.
    - Fins 9y ago
      It seems that GMaps uses different routing engine, but GMaps often says that some traffic accident data is from Waze. With Waze, it almost always tells me to get off the highway at a certain point and then take a parallel (smaller) road to work. If I, as I now almost always do, ignore it and drive past the exit to take one couple miles down (and much closer to work) I can see Waze's ETA dropping by a few minutes. I complained to Waze about it, bit don't see any changes so far.
several 9y ago
I'm glad to see that one of the neighboring towns of Leonia NJ (the town that attempted to restrict the use of public roads to residents only from 6-10AM and 5-9PM) is suing the town over the constitutionality of the restrictions: http://thepressgroup.net/concerns-over-leonia-road-closures-local-residents-only-rule-draws-backlash-lawsuit/ http://thepressgroup.net/concerns-over-leonia-road-closures-...
atomical 9y ago
Has no one reverse engineered the Waze API to add artificial slowdowns in their neighborhood?
- shagie 9y ago
  Yes, they've tried. Traffic-weary homeowners and Waze are at war, again. Guess who’s winning? - https://www.washingtonpost.com/local/traffic-weary-homeowners-and-waze-are-at-war-again-guess-whos-winning/2016/06/05/c466df46-299d-11e6-b989-4e5479715b54_story.html https://www.washingtonpost.com/local/traffic-weary-homeowner... The conclusion to the story: > Soon after, Connor went rogue. He experimented with Waze, confirmed it was sending drivers down his street and began filing his false reports. > “It didn’t do much and within two weeks they stopped showing up on the map all together,” Connor said. “They were on to me.” > The traffic flow began to wane when the road construction ended, Connor said, but remains three or four times higher than before it began. For some drivers, their app-inspired shortcut became a permanent route. HN on the WaPo article from 2 years ago: https://news.ycombinator.com/item?id=11846722 https://news.ycombinator.com/item?id=11846722
  - stergios 9y ago
    There's an easier way to hack waze if you live on a small residential street. All one has to do is buy a $500 dollar car. Then have it "break down" in the middle of the road, essentially blocking the road. Suddenly your car starts up 10 mins later and moves, but after legit waze users report a problem. Repeat until traffic is reduced. You can coordinate with your neighbors to take turns. Civil disobedience can go a long waze to reduce traffic.... I gotta go!
    - largote 9y ago
      Until you find something else to be up-in-arms about.
  - atomical 9y ago
    Can't you use AI to fight AI? Especially if you know the protocol.
quotemstr 9y ago
It's amazing how so many people jump immediately to censorship as a way to solve problems. The idea that Waze can be sued merely for giving people information about routes is absurd, yet here we are.
quotemstr 9y ago
I'm a fan of applying congestion charges to every road, not just arterials. Why not? We have the technology. The price mechanism is what lets us allocate scarce resources to their most efficient use, and applying it to the very short term real estate market for road space can only ensure that we get the maximum economic value out of a fixed number of lane miles.
- jngreenlee 9y ago
  I like your approach, which could also be stated as "either build more if you're going to charge 0, or else charge based on privacy/demand" This would force employers to look at flexible hours, work from home, etc., to keep employees. Ive often thought how strange it is that the marginal use of a road is free...which can cause waste, abuse of the commons, etc. It's also strange to be on a high floor in SF, CHI, NYC, etc., And look down at the grid of streets that could be valuable real estate...yet something like 50% of all and (and the air space above) is dedicated to hundred-year-old technology!
  - quotemstr 9y ago
    Right. "Free" triggers a host of bugs in the human cost estimation system. Put a price on something and we can start to approach it more rationally and make the necessary trade-offs. I suspect funding for transit would increase if the tolls needed to keep city streets at a free-flow traffic level were immediately apparent.
    - b1daly 9y ago
      As a counter argument , forcing people to factor in the cost of a toll as a factor in making an a given trio could just as easily lead to sub-optimal underutilization of what has historically (in the US anyway) a public good. People are no more rational about making small purchases than they are about providing the necessary funds through taxation. I suspect most people intuitively grasp that having freely accessible public roadways is well worth the cost.
      - 9y ago
YokoZar 9y ago
> This description fits Los Angeles pretty well. In 2001, Los Angeles County boasted 21,085 lane miles of maintained highways. In 2016, that number had not budged much, growing to only 21,826 lane miles. In the same period of time, the number of vehicle-miles traveled by Los Angeles commuters rose by some 10 million per day. A major reason why vehicle-miles traveled increased is that constructing new housing in Los Angelos is illegal. People still wanted to join the city, but now they must commute in from further and further away. Congestion is the inevitable result. Housing policy and transportation policy are inextricably tied.
Simulacra 9y ago
I use Waze primarily to avoid police traps, and to give certainty to my commute. If it says I will be there in an hour, it’s usually right. That makes it much easier to plan that commute, for example stopping to get gas or breakfast etc.
saosebastiao 9y ago
Braess' paradox, where increases in road capacity can actually reduce traffic capacity, relies on an important assumption: that people are acting rationally with perfect information. In other words, it is that individually optimal behavior causes the breakdown for the system as a whole. Funnily enough, the blame on Waze might be on point: we might actually be experiencing Braess' paradox more often now, not due to building roads, but because people are more likely to optimize their routes due to traffic-aware GPS systems. And in that respect, Waze could be causing network-wide congestion while simultaneously providing the best possible service for its users. Of course if that were true, it would also mean that you could also blame the road network and that Waze is merely using it as it was designed. Then we get into the the philosophical argument about the drunk jaywalker being hit by a speeding car: You can blame either party for 100% of the blame, 0% of the blame, or arbitrarily apportion blame between the two parties in any way you want, and no matter what, your judgment will be a pure reflection of your biases.
- jaysonelliot 9y ago
  And equally true, the real question is not who was at fault, but how to best solve the problem.
rhacker 9y ago
I lived in Portland for about 4 years (this was 4 years ago). Recently moved back there for a period of 6 months and then moved away again. The entire area changed traffic wise to what I would call insanity. If you use Google Maps or whatever it routes everyone off the main freeways now, and dumps everyone into the farm communities. It's really sad because people don't change their asshole driving behavior. The people in these farms now have to listen to BOOM BOOM BOOM of radios all the time because of the rest of the infrastructure being totally blasted. I personally don't believe Waze is to blame, I'm pretty sure it's google maps (same company) but only Alphabet has the numbers on both.
- tjr225 9y ago
  I see this in the Seattle area as a commuter as well. By the time you've been redirected to the surrounding rural communities it's too late - so have hundreds of other drivers. To pretend that these two lane roads are built for the traffic of I5 IS insanity. Sometimes I see a suggestion from Waze and can't help but laugh at my phone.
  - dte22 9y ago
    Waze does distribute redirected traffic if there are other alternatives. I've seen it first hand on the 401 in Toronto; an accident occurred, I was routed one way to clear it and the person beside me had Waze went another way.. 20km later.. we are both back on the 401! Same car, same person, both using Waze.
    - tjr225 9y ago
      That's one of the difficulties of commuting the I5 corridor. We're stuck between a huge body of water and a mountain range - there really aren't that many alternatives - and the ones that do exist are already heavily trafficked.
sho 9y ago
From the article: > In 2017, the average L.A. commuter spent 102 hours in rush hour traffic, making it the most congested city in the world Having lived in Jakarta and Bangkok, this is extremely hard to believe. LA is horrible for a supposed first world country, sure, but it's nothing like some of the asian megacities. 102 hours a year? I'd wager many jakartans spend 102 hours per month! Very questionable reporting IMO.
- lyqwyd 9y ago
  My suspicion is the statistic is too vague to be meaningful. I would bet that the key here is the average person in LA commutes every day, but I suspect the person in Jakarta does not commute in a way that is easily measured for the purpose of this statistic.
  - sho 9y ago
    Very possibly. They perhaps simply don't have the numbers. But regardless, for anyone who has travelled, the assertion that LA is > the most congested city in the world is simply laughable.
- nxc18 9y ago
  This was a transparent attempt to use techies love of Waze to influence opinion about ultra conservative public planning ideas. Building more roads does not make traffic better. They shit on public transit while completely ignoring any other benefits (e.g. cost efficiency, energy efficiency, safety) or the possibility of growth. It was obvious that this article was bullshit within the first few paragraphs.
1ris 9y ago
>Their bizarre strategy: making those neighborhood streets worse to drive on. What is bizarre about this? This is the empirical approved solution for this: More and better streets have cause more traffic, fewer and worse have less. If you want less trafic this strategy is as good as it gets. And if people want few trafic that what politics should do. I'd guess a website called "reason.com" would like empirical data, and science, but they seem to care about some anecdotes. https://usa.streetsblog.org/2017/06/21/the-science-is-clear-more-highways-equals-more-traffic-why-are-dots-still-ignoring-it/ https://usa.streetsblog.org/2017/06/21/the-science-is-clear-...
- BlackFly 9y ago
  Yeah, the article claims that everything backfires but it seems to me the issue is that they are not using the strategies in tandem. Look at the Netherlands: residential areas are designed so that it is not possible for through traffic to have a shorter route (or a route at all) through them. You may be 100 meters from your school if you walk your kids there, but 3 km if you drive. Then they make a plethora of bike and bus lanes to entice people to drive less and reward people who do not. These strategies work: neighbourhoods are safe, the roads are congested (they always will be) but people have viable alternatives.
jakecopp 9y ago
I wish it was politically possible to make more dedicated bus lanes. When I see a couple of private cars blocking 3 buses with ~100 people in each I get so angry. From a utilitarian point of view bus only lanes are a no-brainer, at least in peak. A visualization: http://1.bp.blogspot.com/-tfJHpDfakqo/Up_mmjALPtI/AAAAAAABQMI/0HglbuZ67Ss/s1600/1450162_455633184542419_1077834738_n%5B1%5D.jpg http://1.bp.blogspot.com/-tfJHpDfakqo/Up_mmjALPtI/AAAAAAABQM...
- closeparen 9y ago
  Consider the empty space between buses on a dedicated bus lane with realistic headways.
  - jakecopp 9y ago
    For a correct comparison, you would have to combine the 5 car traffic lanes into one to compare throughput to the bus lane. My previous image looks like maybe 500m, so say 2.5km long line of cars. Cars can't travel that close together. For driving at say 70kph which is around 20m/s, for a safe 3s distance between cars you need 60m for each car. I guess around 30 cars in each lane of the photo, so 5x30=150 cars 60m x 150 cars = 9000m= 9km So now we are up to a 9+2.5=11.5km line of cars PER BUS!! I would argue you can get better bus headway than that right now, but if a bus has its own lane you get faster average speed, which means you have less buses on the route at any one point, which means you can keep the original number of buses and run them with high frequency/smaller headway. Cars will never be able to compete with public transport with dedicated right of way. If only cars would get out of the damn way of buses and light rail. Feel free to show me I'm wrong though, happy to learn.
    - closeparen 9y ago
      11.5km at 70kph is 10 minutes. That's on the aggressive side (but plausible) for peak-hours frequency of the most popular bus routes in the top 5 American cities for public transit. Just to break even with car traffic. That was my point. 60m following distance is quite optimistic, the bus's stops count for something, and the only way you're getting 150 people on a bus is with the majority as standees, so you're comparing luxury living-room couches to a sardine can, and the sardine can isn't even doing that well on capacity. I agree that in a theoretical universe with i.e. 2-minute bus headways, the bus wins hands down, but I'm not optimistic that any American BRT project will even try for better than 15.
9erdelta 9y ago
In LA all you need is a train from Sherman Oaks into one of the stations in Santa Monica or Culver City. That would cut way down on the 405 congestion. It gets tedious reading articles that come up with "stop funding public transit" as a conclusion. In math class that would be the "trivial" therefore useless answer. Seeing as how cities have more roadway now than ever, yet congestion is arguably getting worse...Clearly, building more roads doesn't work. Because WE HAVE built more roads. The Sepulveda pass in LA is not that old, yet it is so congested as to be useless every single day of the week, at almost every hour of the day (ok maybe this is an exaggeration). One thing I have noticed is Google Maps will suggest going off a main street for a few blocks then merge back into the said main route. This cuts off everyone who's been waiting in line, and potentially puts me in a situation of having to make a left across perhaps 4 lanes of congested traffic, with no light or turn signal. This _IS_ kind of assinine, and it certainly wouldn't hurt for Google/Waze et. al. to feel some pressure to cut these suggestions out.
- cleansy 9y ago
  > Clearly, building more roads doesn't work There is a great video[1] by Wendover Productions about how to fix traffic, explaining exactly this effect. tl;dr: The more road capacity you supply, the more traffic will come since potential demand will "always" outstrip supply. [1]: https://www.youtube.com/watch?v=N4PW66_g6XA https://www.youtube.com/watch?v=N4PW66_g6XA
  - pfranz 9y ago
    Citizens care a lot about travel time (but not enough to live closer, move, or take another mode of transportation), cities and governments care about throughput (economic opportunity and growth). I think often they ignore each other. I'm also curious how those traffic issues apply to subways. I feel like I've seen similar problems where subways exceed capacity. I wonder if the cost/benefit or travel time/capacity is better or worse?
  - ivionday 9y ago
    Related: Jevons Paradox. https://en.m.wikipedia.org/wiki/Jevons_paradox https://en.m.wikipedia.org/wiki/Jevons_paradox
thinkingemote 9y ago
I think Waze uses psychology more than their competitors, and I'll tell you how. They give the appearance of being quicker. Journeys with Waze feel faster. Your brain is more active when driving a Waze route. You remember driving along a Waze route. Most people when they commute enter autopilot mode. They often can't remember the details of what they were doing 20 minutes ago when they arrive at their destination. People commuting often do not need Sat Navs - they will be driving the same way twice a day for months if not years. To change to an active mode of driving where you pay attention and are out of auto pilot makes it feel more faster. Magicians and story tellers will tell you that it's the memory of the performance or story which counts the most - what the user takes with them after is important, and can be shaped. More turns, more intersections. More fast stretches of road (and more stops) makes for a more exhilarating experience driving. It might not be more quicker, or it might be. Most (i.e. not HN readers) people won't notice, or if takes a little bit longer, they will feel that it's quicker. Remember - Waze had gamification built right into the platform. Drivers were encouraged to go out of their normal way, to engage with the platform for years. They know what gamification means at a psychological level. Regular routing considers shortest path, quickest route, simplest paths, stick to highways etc. Waze prioritises driving that makes the user appreciate Waze more.
- gnode 9y ago
  I haven't used Waze, but this is something that I do consciously. I find it's more beneficial for my mental well-being to choose a less direct, longer route, that I can drive continuously on, over a quicker but more congested route with lots of waits in queues and at traffic lights. I'm reminded of a comment I read on HN the other day, which expressed that anger arises from being blocked from progressing by things outside of your control but not outside of possibility. People generally don't care if something takes a long time, they care when they notice reasons why it need not take as long a time as it does.
  - garyrichardson 9y ago
    I also experience this. When I'm actually driving I feel like I'm actually doing something. My blood pressure rises if I'm queued up at a light for 3 cycles. In theory, I want more roundabouts in my area but I don't know if it actually increases flow.
rootlocus 9y ago
> [...] Waze, a navigation app that alerts motorists to alternative routes on residential roads, away from the clogged and congested highways. > Last week City Councilmember David Ryu imploring the city's attorney to take some form of unspecified legal action against the app. So they want to take legal action against an app that tells people they can use public routes which they didn't know about because the locals along those routes don't like the additional traffic? Sounds like they're saying "We don't want you to know you can use this road because we don't like having additional traffic". And they want to take legal action to enforce it. What a joke.
- dfxm12 9y ago
  This is probably 99% NIMBY complaining, but there are legit concerns that 1. Some side street isn't designed to handle the volume of traffic caused by this and is likely to wear out the roads much quicker 2. People coming off the highway might not know they're entering a residential street and don't respect the slower speed limit 3. Increased traffic in a residential neighborhood in general leads to less safe streets, poorer air quality, etc. Of course, the fixes should probably be along the lines of a speed trap, additional public transportation, putting some forethought into city planning, etc. and not a lawsuit.
  - rootlocus 9y ago
    Those are sensible concerns, thank you for pointing them out.
- lolsal 9y ago
  Roads are designed with a capacity and traffic patterns in mind. It's not a joke that these routes are being abused. It's not just NIMBYism.
dandare 9y ago
> In 2001, Los Angeles County boasted 21,085 lane miles of maintained highways. In 2016, that number had not budged much, growing to only 21,826 lane miles. In the same period of time, the number of vehicle-miles traveled by Los Angeles commuters rose by some 10 million per day. Sloppy argumentation. Just becuase 10 million is a large number it does not mean the increase is more than the 3.5% increase in available lane miles.
w-m 9y ago
What a fun read. Interesting to see how often within a single article the author can push the same agenda, without any arguments to support it or objectivity of any kind. Why would you read a magazine full of stuff like that? Quick peek into the finance of the reason foundation: https://www.sourcewatch.org/index.php/Reason_Foundation https://www.sourcewatch.org/index.php/Reason_Foundation And for a good laugh, their headlines on public transport: https://reason.com/tags/public-transportation https://reason.com/tags/public-transportation
- snikeris 9y ago
  What agenda was being pushed in the article? What's your agenda in posting this comment?
  - w-m 9y ago
    Agenda I felt being pushed in the article: Agenda a): building more roads and highways is the only possible solution to congestion issues, either implied or directly called for: > The cause of worsening congestion, says Moore, is pretty simple: more people wanting to drive on the same amount of road. > In 2016, that number had not budged much, growing to only 21,826 lane miles. In the same period of time, the number of vehicle-miles traveled by Los Angeles commuters rose by some 10 million per day. > Doing this, says Moore, requires a mix of road redesign to better handle thru traffic, congestion pricing (whereby drivers pay a variable toll depending on the number of cars on the road on existing road capacity), and building new roads to meet demand. > Some metro areas are already putting these ideas to the test. Washington, D.C., has built what are known as queue duckers... > Ultimately, Moore says, Los Angeles needs to redirect its transportation dollars away from a little-used public transit system—which eats up about half of the city's transit budget—and into building more roads and adding lane-miles to meet increasing demand. Agenda b): reducing car traffic kills local business (several places as well, please go through it if interested). I posted the comment to highlight that this is not a researched journalistic piece, but one-sided article from somebody who makes the same point over and over, completely unsubstantiated. My agenda was to discredit the source as I don't think it's well written enough to be a good starting point for an interesting discussion.
sevensor 9y ago
So a writer for Reason, the supposedly libertarian magazine, comes out in favor of government-funded road building instead of government-funded transit? Way to stand on principle there.
praptak 9y ago
There is a solution for this, although it does not necessarily apply to existing situations, it's from "A Pattern Language: Towns, Buildings, Construction" A Pattern Language: Towns, Buildings, Construction, the architecture book that spawned the Software Design Patterns movement. "49 Looped Local Roads Nobody wants fast through traffic going by their homes. Through traffic is fast, noisy, and dangerous. At the same time cars are important, and cannot be excluded altogether from the areas where people live. Local roads must provide access to houses but prevent traffic from coming through. This problem can only be solved if all roads which have houses on them are laid out to be "loops." We define a looped road as any road in a road network so placed that no path along other roads in the road network can be shortened by travel along the "loop.'' The loops themselves must be designed to discourage high volumes or high speeds: this depends on the total number of houses served by the loop, the road surface, the road width, and the number of curves and corners. [...]"
- organsnyder 9y ago
  This design—which is fairly common in suburban areas—leads to huge headaches when the road is constricted due to construction, accidents, special events, etc. A grid system is much more efficient at adapting to varied conditions.
d0ugie 9y ago
I drive from Connecticut to Manhattan and back during rush hour and have been reliant on Waze's aggressive and inventive routing decisions that often take me through quiet neighborhoods to hop from one highway to another, dodging traffic jams I presume. It almost always gets me there by its eta. On one hand I'm sympathetic to these people (my wife and I are house-shopping and one of our factors is the house being on a quiet road, also I don't particularly like out-of-state drivers in my town, for example New Yorkers don't realize you can turn on red in Connecticut, drives me nuts), on the other hand I can't be late to my shrink sessions - my guy ain't cheap!
- taborj 9y ago
  From someone who lives on a road routinely used to cut about 15 seconds off people's commute (seriously), I have two asks: obey the speed limit, and stop at the stop signs. That's all we're asking. As I'm typing this, another car has blasted through the stop signs at well over the 25mph posted limit. Time to give another call to the local PD.
  - bagacrap 9y ago
    Stop signs are shitty and inefficient traffic calming measures. Call your pd but then hang up and call your city planners and ask for better road features, e.g roundabouts and other physical features https://nyc.streetsblog.org/2011/04/26/to-get-safer-streets-traffic-lights-and-stop-signs-arent-the-answer/ https://nyc.streetsblog.org/2011/04/26/to-get-safer-streets-...
    - taborj 9y ago
      The stupid thing is, we all know it. The city planners didn't want to put stop signs in, but my neighbors demanded it. It was stop signs or speed bumps, and the majority ruled in favor of stop signs (not that speed bumps would have been better). Really, for our street - which goes between a rural highway and a major local artery - the only solutions are to give drivers another more convenient way to cut through, or to shut off the entrance from the highway, forcing them to go another 1/4 mile to the stop light as intended.
mannykannot 9y ago
It used to be that if you could use a map, you could often do significantly better than average, but these days, you are likely to be caught in Waze-directed traffic whatever you do. Nowadays, a route that is longer but less frustrating is usually the best you can hope for, and that not so often.
rhinoceraptor 9y ago
The only ticket I've ever gotten was from taking a Waze shortcut through a neighborhood, where through traffic wasn't allowed.
squozzer 9y ago
In the ATL, a common antidote to residential cut-throughs - where drivers also routinely exceed the posted speed limit - is speed bumps. Not the ones normally seen in shopping center parking lots, but ramps that will send a car airborne when they go too fast. They probably don't work on Dukes of Hazzard fans but on most others they at least slow them down. Another solution would be to ban subdivisions from connecting to two or more larger roads. The downside might be that delivering US Mail - or garbage pick-ups - would become less effcient.
yingbo 9y ago
Waze did nothing wrong here. As long as the road is owned by public, and was built and benign maintained by all tax payers, all of us have the right to drive on it. It's selfish and unfair to even try to stop others driving on them.

134 points by Fins 9y ago

Driving on neighborhood streets is asocial and callous. Wether you use an "app' or not is irrelevant, you're still wrong. As social community feelings have eroded, I guess we now must use technology to combat those that are not polite enough to comply. Make all but the roads designed for transit prohibited to all but local traffic, enforce by cutting streets or ANPR cameras.

seanmcdirmid 9y ago
It’s all good if your roads are privately funded. If they are funded by taxpayers though...
- ghaff 9y ago
  Not everything is solely about legality.
- maxerickson 9y ago
  If they are funded by taxpayers taxpayers can set rules for their use?
  - seanmcdirmid 9y ago
    Yes, as long as the minority doesn’t rule the majority.
c22 9y ago
While I agree with your sentiment, technologically barring residential streets for non-local traffic would have a lot of downsides anytime your friends want to visit or you order a pizza.
- zachkatz 9y ago
  That counts as local traffic
  - c22 9y ago
    Yes, that's my point.
rootlocus 9y ago
> Wether you use an "app' or not is irrelevant, you're still wrong. So I'm wrong to drive on public streets built with public funds because you feel entitled to it by living on close proximity? It might be inconvenient for you, but it's not wrong. > Make all but the roads designed for transit prohibited to all but local traffic As long as the prohibited roads are payed for by the locals, sure. There are plenty of residential complexes with private roads, but don't pretend like public areas are private.

ghaff 9y ago

Waze is actually only one of the topics the article covers. But it’s hard to see what anyone can do. It’s arguably obnoxious to routinely route people on quiet residential streets to save a few minutes. However, they are public streets and experienced commuters might well do the same routing.

GCU-Empiricist 9y ago
So does Waze do something more aggressive than Gooogle's faster route option. This is the second article Ive seen blaming Waze, but I get similar recommendation's as a casual Google Maps user who checks which of three routes is fastest in each morning.
- addicted 9y ago
  I always assumed that Google used Waze data/algorithm for their maps? I would be surprised if they bought Waze, but didn't integrate it with Maps at all.
  - Bedon292 9y ago
    Waze is definitely more aggressive about its neighborhood routing. Though they could be using the same underlying code somewhere, its definitely tuned differently. The main integration point into Google Maps is the accident reporting and active traffic data.
    - sjwright 9y ago
      In my experience, I find that Waze seems to underestimate the time cost of intersections. I think that's why it is more willing to route you through local streets than its competitors.
      - TheAdamist 9y ago
        I find it overestimates traffic lights sometimes, on a medium road next to a grid neighborhood to turn right on another medium road, generally has you cut the corner by one block to avoid the light, even if you can't ever make it out into traffic before the light changes anyway. Try and safely pay attention to wazes route and ignore it sometimes, but it's generally correct. And you can send audio prompts to a phone speaker instead of Bluetooth, unlike gmaps, so that's also useful.

enoch_r 9y ago

As someone who leans quite strongly libertarian, I do not understand what seems to be the default libertarian position on roads. The default view seems to be that government-funded highways are essentially a god-given right. If I want to travel 30 miles every day from home to work, well, somebody had better build me a 8-lane highway to get me there with a minimum of congestion. (Sure, maybe sometimes we'll implement congestion pricing that will cover a miniscules portion of the cost of building and maintaining the highway. This is still a massive government subsidy for people who drive.) Bike lanes, on the other hand? Public transit? Those get the more typical libertarian viewpoint on government spending. How dare the government try to distort our incentives by spending millions of dollars to provide bike infrastructure?! (Fun fact: As of 2008, the replacement cost for all of Portland's bike infrastructure was estimated at $60M, the rough cost of one mile of urban freeway. I'm sure it's gone up by then--it's still a rounding error.) It's genuinely weird how libertarians--generally not ones to ignore the effects of government spending on incentives--don't appear to notice that spending hundreds of billions of dollars on a transportation system that you can only use if you're in a car might actually, you know, cause more people to drive.

Analemma_ 9y ago
> The default view seems to be that government-funded highways are essentially a god-given right. What sort of libertarians are you hanging out with? I'm not a libertarian myself, but most of the ones I have talked to about roads wouldn't agree with this. Their usual position is that roads should be privatized and tolled, and that this would help fix the externalities (traffic, concrete deserts, suburban sprawl) caused by "free" access to roads, if people had to pay their true cost.
- acdha 9y ago
  I think this is an example of real versus expressed preferences: I’ve known a bunch of libertarians who would when thinking about the issue say all roads should be toll roads but at all other times act as if they should be taxpayer funded, complain that their taxes are way too high and should pay for plenty of roads, talk like construction prices shouldn’t have gone up since 1960, etc. Their unconscious way of thinking about it was basically the same as everyone else, and I think that’s in part because like everyone else they didn’t really want to pay every time they drive somewhere.

hprotagonist 9y ago

I remember reading some years ago about a village in the UK that wanted to be removed from satnav maps. At that time, most of the people who used them were truck drivers, and the 12th-century village's main road was about 3 inches too narrow to allow a truck through. A wedged truck on main street a week for 8 months, and people got pretty cranky about it. I'm not sure what ever happened, though.

reificator 9y ago
> A wedged truck on main street a week for 8 months, and people got pretty cranky about it. I think you accidentally a word there.
- ocdtrekkie 9y ago
  Not sure it's actually wrong, but it does read weird. "A wedged truck on main street [each] week for eight months" reads a little better, but I don't think what the parent comment said was technically incorrectly written.
  - hprotagonist 9y ago
    Perhaps I should’ve said “One truck” instead of “A truck”.
    - ocdtrekkie 9y ago
      That would've read a little easier as well.
      - hprotagonist 9y ago
        the fancy-pants term for this, i think, is https://en.wikipedia.org/wiki/Garden_path_sentence https://en.wikipedia.org/wiki/Garden_path_sentence
- lainga 9y ago
  It's a clipped version of After putting up with [one] wedged truck..., it seems valid to me. Similar to An apple a day keeps the doctor away, or more similarly, An apple a day, and the doctor stays away.

ndaiger 9y ago

I live in Los Angeles near a lot of the construction going on for the new subway line running down Wilshire Blvd. My neighborhood (Carthay) has been getting a lot more cut-through traffic. There are days that seem particularly "Waze-y", when it's basically just gridlock on our little streets for an hour. I imagine it's due to apps re-routing people. But I don't really think it's any faster for the commuter. I'm reminded of this article (seemingly now password-protected, so I'll link to a summary) that Waze chronically underestimates your travel time, while Apple Maps chronically overestimates (and therefore overdelivers): https://appleinsider.com/articles/18/02/22/informal-testing-finds-apple-maps-arrival-times-are-intentionally-conservative-to-provide-a-good-user-experience https://appleinsider.com/articles/18/02/22/informal-testing-... From my anecdotal experience, Waze does vastly overestimate the benefits of a crazy route, like making a left turn from a stop sign across 6 lanes of rush-hour traffic.

ghaff 9y ago
Google Maps does this too at times. It isn’t so much residential streets but I’ve had Google take me on windy country drives where I inevitably screw up.
wyclif 9y ago
It's ludicrous that Los Angeles politicians are blaming Waze for this, when it's their own fault for failing to approve and build higher-volume and better roads. I smirk every time I read some bureaucrat blaming apps that demonstrably make our lives better instead of owning up to their own impotence and incompetence.
- bsagdiyev 9y ago
  Have you been to Los Angeles? On what space do you expect them to do this on, without interrupting the already bad traffic and with what money? I'm genuinely curious as to your idea and how it would work.
  - wpietri 9y ago
    And even if they somehow do manage to do it, I'm not sure at all that any lasting improvement will be made. An urban planner friend tells me that people decide where to live based on commute time, not miles driven. So if you put in more roadway, people move to cheaper places farther out, using more road-miles than before, at least until the speeds drop back down.

jcranmer 9y ago

For city planners, there's a natural hierarchy of roads: limited-access freeways, arterials, collectors, and residential streets. Moving high volumes of cars is intended to use arterials and freeways, whereas collectors and residential streets are deliberately designed for lower speeds as they may have a higher likelihood of pedestrians randomly entering the street (the child racing after his ball, to use the proverbial example). In a nutshell, the problem with Waze is that it encourages drivers to shift from arterials to collector roads, where they very often proceed to drive in flagrant violation of safety standards. Getting drivers to not do that is challenging (and this problem predates Waze). Speed bumps don't meaningfully slow people down. Even prohibiting the through movement in a T intersection by means of an island doesn't work--people will happily drive over the curb to do so. By routing people down roads that are not designed for this traffic, Waze is pretty close to being complicit in these safety violations. The proximate cause of this furor is that there is a street in LA that is too steep to be safe for through traffic, yet Waze is happily routing people along the street. Waze's response is, effectively, it's legal for through traffic, so Waze doesn't have to do anything, with the underlying reasoning being that it doesn't want to have to deal with everyone demanding special treatment. Waze's attitude is, quite frankly, irresponsible. What they could do--and should be doing--is to look at potential factors that impact safety and update they app to not route people on roads that are clearly unsafe (Google's maps already have the ability to report grade on roads, for example). In fact, that's what the LA council is basically doing: they're looking to see if Waze could be doing more, and in the likely case they find that Waze is indeed not doing enough, finding some mechanism to compel Waze to do more. The other element in play here is a complete about congestion. The truth is... you can't fix congestion. In transportation theory, there's an element called induced demand: if travel time improves in a corridor, then that induces people to take extra trips, often returning it fairly quickly to the same level of congestion. What many cities have realized is that trying to provide sufficient road capacity to meet demand is a game lost before they've even begun, and instead focused their efforts on trying to coax better utilization of current corridors (single-occupant vehicles on roads are basically the lowest-throughput transportation measures devised, except for Elon Musk's proposals which somehow manage to be even worse).

galdosdi 9y ago

CarVac 9y ago

I live in Leonia, as mentioned in the article, and while I'm unsure of how much less traffic businesses are actually seeing, I do notice that when I'm navigating to places out of town, Google Maps and Waze never route me on local streets inside town anymore, preferring to go on the main thoroughfares until outside the town, and then crossing between them on other towns' local streets.

dkural 9y ago

It's ironic that the article seems to think building more roads, thereby improving road-to-miles-driven ratio will ease congestion. It also takes as fact that it's "as inefficient" to use Waze. My experience is Waze often does cut commute time down; not always; but on the aggregate it demonstrably, scientifically, provably does. That's why it's a huge success and people use it. I do think there are problems with drivers disregarding the law (speed limits, signs etc.) on smaller streets, and perhaps they need to be more aggressively punished for doing so. In general I am for more aggressive fees, sentencing and punishment for violating driving rules.

steve_musk 9y ago
Where is the scientific proof? Legitimately curious.
Fins 9y ago
But did you have somebody leaving with you, and NOT using Waze to prove that Waze does cut commute down compared to something else? Disclaimer: I usually do use Waze commuting, but find its suggestions in certain places downright bizarre.
- Casseres 9y ago
  Waze is now owned by Google, but I don't know if they share the same route-suggestion engine. I've had Google Maps (desktop web browser) give me a route once through heavy traffic, but when I dragged the route to a mostly-parallel highway, my estimated time decreased by 10 minutes. So it's not perfect, or they wanted more users to give more data on the traffic.
  - Fins 9y ago
    It seems that GMaps uses different routing engine, but GMaps often says that some traffic accident data is from Waze. With Waze, it almost always tells me to get off the highway at a certain point and then take a parallel (smaller) road to work. If I, as I now almost always do, ignore it and drive past the exit to take one couple miles down (and much closer to work) I can see Waze's ETA dropping by a few minutes. I complained to Waze about it, bit don't see any changes so far.

several 9y ago

I'm glad to see that one of the neighboring towns of Leonia NJ (the town that attempted to restrict the use of public roads to residents only from 6-10AM and 5-9PM) is suing the town over the constitutionality of the restrictions: http://thepressgroup.net/concerns-over-leonia-road-closures-local-residents-only-rule-draws-backlash-lawsuit/ http://thepressgroup.net/concerns-over-leonia-road-closures-...

atomical 9y ago

Has no one reverse engineered the Waze API to add artificial slowdowns in their neighborhood?

shagie 9y ago
Yes, they've tried. Traffic-weary homeowners and Waze are at war, again. Guess who’s winning? - https://www.washingtonpost.com/local/traffic-weary-homeowners-and-waze-are-at-war-again-guess-whos-winning/2016/06/05/c466df46-299d-11e6-b989-4e5479715b54_story.html https://www.washingtonpost.com/local/traffic-weary-homeowner... The conclusion to the story: > Soon after, Connor went rogue. He experimented with Waze, confirmed it was sending drivers down his street and began filing his false reports. > “It didn’t do much and within two weeks they stopped showing up on the map all together,” Connor said. “They were on to me.” > The traffic flow began to wane when the road construction ended, Connor said, but remains three or four times higher than before it began. For some drivers, their app-inspired shortcut became a permanent route. HN on the WaPo article from 2 years ago: https://news.ycombinator.com/item?id=11846722 https://news.ycombinator.com/item?id=11846722
- stergios 9y ago
  There's an easier way to hack waze if you live on a small residential street. All one has to do is buy a $500 dollar car. Then have it "break down" in the middle of the road, essentially blocking the road. Suddenly your car starts up 10 mins later and moves, but after legit waze users report a problem. Repeat until traffic is reduced. You can coordinate with your neighbors to take turns. Civil disobedience can go a long waze to reduce traffic.... I gotta go!
  - largote 9y ago
    Until you find something else to be up-in-arms about.
- atomical 9y ago
  Can't you use AI to fight AI? Especially if you know the protocol.

quotemstr 9y ago

It's amazing how so many people jump immediately to censorship as a way to solve problems. The idea that Waze can be sued merely for giving people information about routes is absurd, yet here we are.

quotemstr 9y ago

I'm a fan of applying congestion charges to every road, not just arterials. Why not? We have the technology. The price mechanism is what lets us allocate scarce resources to their most efficient use, and applying it to the very short term real estate market for road space can only ensure that we get the maximum economic value out of a fixed number of lane miles.

jngreenlee 9y ago
I like your approach, which could also be stated as "either build more if you're going to charge 0, or else charge based on privacy/demand" This would force employers to look at flexible hours, work from home, etc., to keep employees. Ive often thought how strange it is that the marginal use of a road is free...which can cause waste, abuse of the commons, etc. It's also strange to be on a high floor in SF, CHI, NYC, etc., And look down at the grid of streets that could be valuable real estate...yet something like 50% of all and (and the air space above) is dedicated to hundred-year-old technology!
- quotemstr 9y ago
  Right. "Free" triggers a host of bugs in the human cost estimation system. Put a price on something and we can start to approach it more rationally and make the necessary trade-offs. I suspect funding for transit would increase if the tolls needed to keep city streets at a free-flow traffic level were immediately apparent.
  - b1daly 9y ago
    As a counter argument , forcing people to factor in the cost of a toll as a factor in making an a given trio could just as easily lead to sub-optimal underutilization of what has historically (in the US anyway) a public good. People are no more rational about making small purchases than they are about providing the necessary funds through taxation. I suspect most people intuitively grasp that having freely accessible public roadways is well worth the cost.
    - 9y ago

YokoZar 9y ago

> This description fits Los Angeles pretty well. In 2001, Los Angeles County boasted 21,085 lane miles of maintained highways. In 2016, that number had not budged much, growing to only 21,826 lane miles. In the same period of time, the number of vehicle-miles traveled by Los Angeles commuters rose by some 10 million per day. A major reason why vehicle-miles traveled increased is that constructing new housing in Los Angelos is illegal. People still wanted to join the city, but now they must commute in from further and further away. Congestion is the inevitable result. Housing policy and transportation policy are inextricably tied.

Simulacra 9y ago

I use Waze primarily to avoid police traps, and to give certainty to my commute. If it says I will be there in an hour, it’s usually right. That makes it much easier to plan that commute, for example stopping to get gas or breakfast etc.

saosebastiao 9y ago

Braess' paradox, where increases in road capacity can actually reduce traffic capacity, relies on an important assumption: that people are acting rationally with perfect information. In other words, it is that individually optimal behavior causes the breakdown for the system as a whole. Funnily enough, the blame on Waze might be on point: we might actually be experiencing Braess' paradox more often now, not due to building roads, but because people are more likely to optimize their routes due to traffic-aware GPS systems. And in that respect, Waze could be causing network-wide congestion while simultaneously providing the best possible service for its users. Of course if that were true, it would also mean that you could also blame the road network and that Waze is merely using it as it was designed. Then we get into the the philosophical argument about the drunk jaywalker being hit by a speeding car: You can blame either party for 100% of the blame, 0% of the blame, or arbitrarily apportion blame between the two parties in any way you want, and no matter what, your judgment will be a pure reflection of your biases.

jaysonelliot 9y ago
And equally true, the real question is not who was at fault, but how to best solve the problem.

rhacker 9y ago

I lived in Portland for about 4 years (this was 4 years ago). Recently moved back there for a period of 6 months and then moved away again. The entire area changed traffic wise to what I would call insanity. If you use Google Maps or whatever it routes everyone off the main freeways now, and dumps everyone into the farm communities. It's really sad because people don't change their asshole driving behavior. The people in these farms now have to listen to BOOM BOOM BOOM of radios all the time because of the rest of the infrastructure being totally blasted. I personally don't believe Waze is to blame, I'm pretty sure it's google maps (same company) but only Alphabet has the numbers on both.

tjr225 9y ago
I see this in the Seattle area as a commuter as well. By the time you've been redirected to the surrounding rural communities it's too late - so have hundreds of other drivers. To pretend that these two lane roads are built for the traffic of I5 IS insanity. Sometimes I see a suggestion from Waze and can't help but laugh at my phone.
- dte22 9y ago
  Waze does distribute redirected traffic if there are other alternatives. I've seen it first hand on the 401 in Toronto; an accident occurred, I was routed one way to clear it and the person beside me had Waze went another way.. 20km later.. we are both back on the 401! Same car, same person, both using Waze.
  - tjr225 9y ago
    That's one of the difficulties of commuting the I5 corridor. We're stuck between a huge body of water and a mountain range - there really aren't that many alternatives - and the ones that do exist are already heavily trafficked.

sho 9y ago

From the article: > In 2017, the average L.A. commuter spent 102 hours in rush hour traffic, making it the most congested city in the world Having lived in Jakarta and Bangkok, this is extremely hard to believe. LA is horrible for a supposed first world country, sure, but it's nothing like some of the asian megacities. 102 hours a year? I'd wager many jakartans spend 102 hours per month! Very questionable reporting IMO.

lyqwyd 9y ago
My suspicion is the statistic is too vague to be meaningful. I would bet that the key here is the average person in LA commutes every day, but I suspect the person in Jakarta does not commute in a way that is easily measured for the purpose of this statistic.
- sho 9y ago
  Very possibly. They perhaps simply don't have the numbers. But regardless, for anyone who has travelled, the assertion that LA is > the most congested city in the world is simply laughable.
nxc18 9y ago
This was a transparent attempt to use techies love of Waze to influence opinion about ultra conservative public planning ideas. Building more roads does not make traffic better. They shit on public transit while completely ignoring any other benefits (e.g. cost efficiency, energy efficiency, safety) or the possibility of growth. It was obvious that this article was bullshit within the first few paragraphs.

1ris 9y ago

>Their bizarre strategy: making those neighborhood streets worse to drive on. What is bizarre about this? This is the empirical approved solution for this: More and better streets have cause more traffic, fewer and worse have less. If you want less trafic this strategy is as good as it gets. And if people want few trafic that what politics should do. I'd guess a website called "reason.com" would like empirical data, and science, but they seem to care about some anecdotes. https://usa.streetsblog.org/2017/06/21/the-science-is-clear-more-highways-equals-more-traffic-why-are-dots-still-ignoring-it/ https://usa.streetsblog.org/2017/06/21/the-science-is-clear-...

BlackFly 9y ago
Yeah, the article claims that everything backfires but it seems to me the issue is that they are not using the strategies in tandem. Look at the Netherlands: residential areas are designed so that it is not possible for through traffic to have a shorter route (or a route at all) through them. You may be 100 meters from your school if you walk your kids there, but 3 km if you drive. Then they make a plethora of bike and bus lanes to entice people to drive less and reward people who do not. These strategies work: neighbourhoods are safe, the roads are congested (they always will be) but people have viable alternatives.

jakecopp 9y ago

I wish it was politically possible to make more dedicated bus lanes. When I see a couple of private cars blocking 3 buses with ~100 people in each I get so angry. From a utilitarian point of view bus only lanes are a no-brainer, at least in peak. A visualization: http://1.bp.blogspot.com/-tfJHpDfakqo/Up_mmjALPtI/AAAAAAABQMI/0HglbuZ67Ss/s1600/1450162_455633184542419_1077834738_n%5B1%5D.jpg http://1.bp.blogspot.com/-tfJHpDfakqo/Up_mmjALPtI/AAAAAAABQM...

closeparen 9y ago
Consider the empty space between buses on a dedicated bus lane with realistic headways.
- jakecopp 9y ago
  For a correct comparison, you would have to combine the 5 car traffic lanes into one to compare throughput to the bus lane. My previous image looks like maybe 500m, so say 2.5km long line of cars. Cars can't travel that close together. For driving at say 70kph which is around 20m/s, for a safe 3s distance between cars you need 60m for each car. I guess around 30 cars in each lane of the photo, so 5x30=150 cars 60m x 150 cars = 9000m= 9km So now we are up to a 9+2.5=11.5km line of cars PER BUS!! I would argue you can get better bus headway than that right now, but if a bus has its own lane you get faster average speed, which means you have less buses on the route at any one point, which means you can keep the original number of buses and run them with high frequency/smaller headway. Cars will never be able to compete with public transport with dedicated right of way. If only cars would get out of the damn way of buses and light rail. Feel free to show me I'm wrong though, happy to learn.
  - closeparen 9y ago
    11.5km at 70kph is 10 minutes. That's on the aggressive side (but plausible) for peak-hours frequency of the most popular bus routes in the top 5 American cities for public transit. Just to break even with car traffic. That was my point. 60m following distance is quite optimistic, the bus's stops count for something, and the only way you're getting 150 people on a bus is with the majority as standees, so you're comparing luxury living-room couches to a sardine can, and the sardine can isn't even doing that well on capacity. I agree that in a theoretical universe with i.e. 2-minute bus headways, the bus wins hands down, but I'm not optimistic that any American BRT project will even try for better than 15.

9erdelta 9y ago

In LA all you need is a train from Sherman Oaks into one of the stations in Santa Monica or Culver City. That would cut way down on the 405 congestion. It gets tedious reading articles that come up with "stop funding public transit" as a conclusion. In math class that would be the "trivial" therefore useless answer. Seeing as how cities have more roadway now than ever, yet congestion is arguably getting worse...Clearly, building more roads doesn't work. Because WE HAVE built more roads. The Sepulveda pass in LA is not that old, yet it is so congested as to be useless every single day of the week, at almost every hour of the day (ok maybe this is an exaggeration). One thing I have noticed is Google Maps will suggest going off a main street for a few blocks then merge back into the said main route. This cuts off everyone who's been waiting in line, and potentially puts me in a situation of having to make a left across perhaps 4 lanes of congested traffic, with no light or turn signal. This _IS_ kind of assinine, and it certainly wouldn't hurt for Google/Waze et. al. to feel some pressure to cut these suggestions out.

cleansy 9y ago
> Clearly, building more roads doesn't work There is a great video[1] by Wendover Productions about how to fix traffic, explaining exactly this effect. tl;dr: The more road capacity you supply, the more traffic will come since potential demand will "always" outstrip supply. [1]: https://www.youtube.com/watch?v=N4PW66_g6XA https://www.youtube.com/watch?v=N4PW66_g6XA
- pfranz 9y ago
  Citizens care a lot about travel time (but not enough to live closer, move, or take another mode of transportation), cities and governments care about throughput (economic opportunity and growth). I think often they ignore each other. I'm also curious how those traffic issues apply to subways. I feel like I've seen similar problems where subways exceed capacity. I wonder if the cost/benefit or travel time/capacity is better or worse?
- ivionday 9y ago
  Related: Jevons Paradox. https://en.m.wikipedia.org/wiki/Jevons_paradox https://en.m.wikipedia.org/wiki/Jevons_paradox

thinkingemote 9y ago

I think Waze uses psychology more than their competitors, and I'll tell you how. They give the appearance of being quicker. Journeys with Waze feel faster. Your brain is more active when driving a Waze route. You remember driving along a Waze route. Most people when they commute enter autopilot mode. They often can't remember the details of what they were doing 20 minutes ago when they arrive at their destination. People commuting often do not need Sat Navs - they will be driving the same way twice a day for months if not years. To change to an active mode of driving where you pay attention and are out of auto pilot makes it feel more faster. Magicians and story tellers will tell you that it's the memory of the performance or story which counts the most - what the user takes with them after is important, and can be shaped. More turns, more intersections. More fast stretches of road (and more stops) makes for a more exhilarating experience driving. It might not be more quicker, or it might be. Most (i.e. not HN readers) people won't notice, or if takes a little bit longer, they will feel that it's quicker. Remember - Waze had gamification built right into the platform. Drivers were encouraged to go out of their normal way, to engage with the platform for years. They know what gamification means at a psychological level. Regular routing considers shortest path, quickest route, simplest paths, stick to highways etc. Waze prioritises driving that makes the user appreciate Waze more.

gnode 9y ago
I haven't used Waze, but this is something that I do consciously. I find it's more beneficial for my mental well-being to choose a less direct, longer route, that I can drive continuously on, over a quicker but more congested route with lots of waits in queues and at traffic lights. I'm reminded of a comment I read on HN the other day, which expressed that anger arises from being blocked from progressing by things outside of your control but not outside of possibility. People generally don't care if something takes a long time, they care when they notice reasons why it need not take as long a time as it does.
- garyrichardson 9y ago
  I also experience this. When I'm actually driving I feel like I'm actually doing something. My blood pressure rises if I'm queued up at a light for 3 cycles. In theory, I want more roundabouts in my area but I don't know if it actually increases flow.

rootlocus 9y ago

> [...] Waze, a navigation app that alerts motorists to alternative routes on residential roads, away from the clogged and congested highways. > Last week City Councilmember David Ryu imploring the city's attorney to take some form of unspecified legal action against the app. So they want to take legal action against an app that tells people they can use public routes which they didn't know about because the locals along those routes don't like the additional traffic? Sounds like they're saying "We don't want you to know you can use this road because we don't like having additional traffic". And they want to take legal action to enforce it. What a joke.

dfxm12 9y ago
This is probably 99% NIMBY complaining, but there are legit concerns that 1. Some side street isn't designed to handle the volume of traffic caused by this and is likely to wear out the roads much quicker 2. People coming off the highway might not know they're entering a residential street and don't respect the slower speed limit 3. Increased traffic in a residential neighborhood in general leads to less safe streets, poorer air quality, etc. Of course, the fixes should probably be along the lines of a speed trap, additional public transportation, putting some forethought into city planning, etc. and not a lawsuit.
- rootlocus 9y ago
  Those are sensible concerns, thank you for pointing them out.
lolsal 9y ago
Roads are designed with a capacity and traffic patterns in mind. It's not a joke that these routes are being abused. It's not just NIMBYism.

dandare 9y ago

> In 2001, Los Angeles County boasted 21,085 lane miles of maintained highways. In 2016, that number had not budged much, growing to only 21,826 lane miles. In the same period of time, the number of vehicle-miles traveled by Los Angeles commuters rose by some 10 million per day. Sloppy argumentation. Just becuase 10 million is a large number it does not mean the increase is more than the 3.5% increase in available lane miles.

w-m 9y ago

What a fun read. Interesting to see how often within a single article the author can push the same agenda, without any arguments to support it or objectivity of any kind. Why would you read a magazine full of stuff like that? Quick peek into the finance of the reason foundation: https://www.sourcewatch.org/index.php/Reason_Foundation https://www.sourcewatch.org/index.php/Reason_Foundation And for a good laugh, their headlines on public transport: https://reason.com/tags/public-transportation https://reason.com/tags/public-transportation

snikeris 9y ago
What agenda was being pushed in the article? What's your agenda in posting this comment?
- w-m 9y ago
  Agenda I felt being pushed in the article: Agenda a): building more roads and highways is the only possible solution to congestion issues, either implied or directly called for: > The cause of worsening congestion, says Moore, is pretty simple: more people wanting to drive on the same amount of road. > In 2016, that number had not budged much, growing to only 21,826 lane miles. In the same period of time, the number of vehicle-miles traveled by Los Angeles commuters rose by some 10 million per day. > Doing this, says Moore, requires a mix of road redesign to better handle thru traffic, congestion pricing (whereby drivers pay a variable toll depending on the number of cars on the road on existing road capacity), and building new roads to meet demand. > Some metro areas are already putting these ideas to the test. Washington, D.C., has built what are known as queue duckers... > Ultimately, Moore says, Los Angeles needs to redirect its transportation dollars away from a little-used public transit system—which eats up about half of the city's transit budget—and into building more roads and adding lane-miles to meet increasing demand. Agenda b): reducing car traffic kills local business (several places as well, please go through it if interested). I posted the comment to highlight that this is not a researched journalistic piece, but one-sided article from somebody who makes the same point over and over, completely unsubstantiated. My agenda was to discredit the source as I don't think it's well written enough to be a good starting point for an interesting discussion.

sevensor 9y ago

So a writer for Reason, the supposedly libertarian magazine, comes out in favor of government-funded road building instead of government-funded transit? Way to stand on principle there.

praptak 9y ago

There is a solution for this, although it does not necessarily apply to existing situations, it's from "A Pattern Language: Towns, Buildings, Construction" A Pattern Language: Towns, Buildings, Construction, the architecture book that spawned the Software Design Patterns movement. "49 Looped Local Roads Nobody wants fast through traffic going by their homes. Through traffic is fast, noisy, and dangerous. At the same time cars are important, and cannot be excluded altogether from the areas where people live. Local roads must provide access to houses but prevent traffic from coming through. This problem can only be solved if all roads which have houses on them are laid out to be "loops." We define a looped road as any road in a road network so placed that no path along other roads in the road network can be shortened by travel along the "loop.'' The loops themselves must be designed to discourage high volumes or high speeds: this depends on the total number of houses served by the loop, the road surface, the road width, and the number of curves and corners. [...]"

organsnyder 9y ago
This design—which is fairly common in suburban areas—leads to huge headaches when the road is constricted due to construction, accidents, special events, etc. A grid system is much more efficient at adapting to varied conditions.

d0ugie 9y ago

I drive from Connecticut to Manhattan and back during rush hour and have been reliant on Waze's aggressive and inventive routing decisions that often take me through quiet neighborhoods to hop from one highway to another, dodging traffic jams I presume. It almost always gets me there by its eta. On one hand I'm sympathetic to these people (my wife and I are house-shopping and one of our factors is the house being on a quiet road, also I don't particularly like out-of-state drivers in my town, for example New Yorkers don't realize you can turn on red in Connecticut, drives me nuts), on the other hand I can't be late to my shrink sessions - my guy ain't cheap!

taborj 9y ago
From someone who lives on a road routinely used to cut about 15 seconds off people's commute (seriously), I have two asks: obey the speed limit, and stop at the stop signs. That's all we're asking. As I'm typing this, another car has blasted through the stop signs at well over the 25mph posted limit. Time to give another call to the local PD.
- bagacrap 9y ago
  Stop signs are shitty and inefficient traffic calming measures. Call your pd but then hang up and call your city planners and ask for better road features, e.g roundabouts and other physical features https://nyc.streetsblog.org/2011/04/26/to-get-safer-streets-traffic-lights-and-stop-signs-arent-the-answer/ https://nyc.streetsblog.org/2011/04/26/to-get-safer-streets-...
  - taborj 9y ago
    The stupid thing is, we all know it. The city planners didn't want to put stop signs in, but my neighbors demanded it. It was stop signs or speed bumps, and the majority ruled in favor of stop signs (not that speed bumps would have been better). Really, for our street - which goes between a rural highway and a major local artery - the only solutions are to give drivers another more convenient way to cut through, or to shut off the entrance from the highway, forcing them to go another 1/4 mile to the stop light as intended.

mannykannot 9y ago

It used to be that if you could use a map, you could often do significantly better than average, but these days, you are likely to be caught in Waze-directed traffic whatever you do. Nowadays, a route that is longer but less frustrating is usually the best you can hope for, and that not so often.

rhinoceraptor 9y ago

The only ticket I've ever gotten was from taking a Waze shortcut through a neighborhood, where through traffic wasn't allowed.

squozzer 9y ago

In the ATL, a common antidote to residential cut-throughs - where drivers also routinely exceed the posted speed limit - is speed bumps. Not the ones normally seen in shopping center parking lots, but ramps that will send a car airborne when they go too fast. They probably don't work on Dukes of Hazzard fans but on most others they at least slow them down. Another solution would be to ban subdivisions from connecting to two or more larger roads. The downside might be that delivering US Mail - or garbage pick-ups - would become less effcient.

yingbo 9y ago

Waze did nothing wrong here. As long as the road is owned by public, and was built and benign maintained by all tax payers, all of us have the right to drive on it. It's selfish and unfair to even try to stop others driving on them.