Today’s Headlines

    • NextBus Gets Arrival Predictions Wrong (SFExaminer)
    • Will Young People Ride the Bus? (MercNews)
    • Vote Coming on Planned Potrero Hill Development (Socketsite)
    • SFPD’s Kohrs Pleads “Not Guilty” to Hit-and-Run Charges (SFExaminer)
    • NFL Pressures Cash-Strapped Cities to Help Buy them a Shiny New Stadium (InsideBayArea)
    • East Bay Nurse Killed by Suspected Drunk Driver Remembered (InsideBayArea)
    • Uber Lampoons Driver Fingerprinting Proposal (InsideBayArea)
    • New Tech May Help Cops Test for “Driving While Stoned” (CBSLocal)
    • Review of “Bikes vs. Cars” Documentary (SFWeekly)

Get national headlines at Streetsblog USA
Get state headlines at Streetsblog CA

  • Jeffrey Baker

    No surprise on the NextBus conclusion. I did a study 7 years ago on the 1-California and the NextBus prediction times for the Sacramento at Fillmore stop were not just inaccurate, but totally uncorrelated with actual arrival time. The service is completely useless.

  • mx

    But Muni just put out a press release about their survey claiming riders think they’re so great! How can the same old things still be broken for so many years?

  • david vartanoff

    Completely useless, no, highly inaccurate, yes. The major issue seems to be that a bus which is scheduled but not responding is shown as operating when it is not. The “ghost buses”

  • Jeffrey Baker

    It is equally likely to have a bus that’s operating but doesn’t register on NextBus, such that NextBus says it will be an hour, but the bus comes 5 minutes later.

    When I said useless I really meant it. At least for the stop I studied you would have done just as well to randomly select any arrival time between zero and 2x the scheduled interval. NextBus was just as bad as a dice roll.

  • murphstahoe

    For an arrival 20+ minutes out, how can that be surprising? Even if there were no traffic, there could be a car accident, a double parker, etc… that would add minutes to the travel time.

  • mx

    I do like this quote in the article: “NextBus sent the following
    statement in response to the study: “In urban environments the time
    between buses may only be 8 to 20 minutes, so people aren’t looking out
    30 minutes for a bus. They want to see the next one or two.””

    Because no one in SF has ever gone to NextBus to be told the next bus is 45 minutes away, right? People don’t want to be looking out 30 minutes for a bus, but that’s what they get.

  • murphstahoe

    SMART comes to compromise over bike paths after MCBC threatens lawsuit.

    http://www.pressdemocrat.com/news/4848365-181/smart-settlement-paves-way-for?artslide=0

    The stretch of path from downtown to the cal-park tunnel would be super useful.

  • joechoj

    NextBus is similarly inept in Oakland. When waiting for AC Transit a couple weeks ago, I checked NextBus to determine frequency – about 15 mins. Deciding it would be worth the wait to ditch the car, my family walked to the bus stop. The bus that was supposedly 8 minutes away never materialized, even though the data feed kept counting down. It got to 0, and then – nothing. We waited another two full cycles of 15 minute countdowns without a sign of a bus. I eventually ran home, grabbed the car, and picked my family up.
    I get that buses are delayed occasionally. But then why does the data feed continue to count down, as though all is well? Clearly, the data feed is completely independent of bus locations. Such a poor execution, and as a result, it’s now that much harder convincing my wife to take the bus. There are real-world consequences to your ridership, transit agencies – wake up!

  • Jeffrey Baker

    If you are near the beginning of the line, NextBus simply counts down to scheduled departures. They have no idea if there’s a vehicle coming or not.

  • Jeffrey Baker

    Sure. The line I’m talking about runs on a 5-minute headway, which poses the opposite problem: if the bus is coming in 5 minutes, do you really need NextBus? But we’re talking about Muni, where “5-minute headway” often means “12 buses nose-to-tail, once an hour.”

    I guess all I ever really wanted from NextBus was an estimate of how screwed up Muni operations are at the moment, on a DEFCON-like scale.

  • joechoj

    > They have no idea if there’s a vehicle coming or not.

    That much is clear 🙂

  • Jeffrey Baker

    Speaking of SMART, does anyone know what happened with their signals yesterday? They announced, vaguely, that:

    “The railroad crossing signal system malfunctioned yesterday on 3rd Street in Santa Rosa. This was due to an issue with the fiber optics cable which is a part of the signal system. Today, SMART is working to fix the problem.”

    Without explaining what happened.

  • CamBam415

    Jeff – That is a good point. Still, transit agencies know if a driver called in sick for a shift, hasn’t clocked in, is a bus is cancelled, etc. I don’t know the specifics, but I’d guess a driver has to be onsite 30-60 min before departing the bus yard? Thus, a transit agency should be able to update NextBus and know that a bus won’t run at a given time (I am sure there are the occasional exceptions).

    From experiences, I used to ride the N/T lines from AT&T Park to the Embarcadero. And had several similar experiences to Joe, where the countdown would show a train is coming and it never materialized. So frustrating as had I been given good info, it would’ve been faster to simply walk. Oh well.

    I don’t expect NextBus will ever be perfect, but there is plenty of room to improve.

  • Jeffrey Baker

    Your complaint is that Muni is operated by goldbricking layabouts, which NextBus cannot solve. Muni operators are entitled to depart the terminal whenever they feel like it. ILEC Park is only 1 stop from the end of the line at 4th & King, so the best NextBus can do is assume the train will leave at the scheduled time, which never happens.

  • CamBam415

    Do you really trust SMART to build that pathway? I don’t. That said, it seems like a little progress has been made, but Huffman and the MTC need to hold SMART accountable, the cycling community just doesn’t have the necessary leverage.

    And agree that is a critical connector. I used to ride from Terra Linda to the Larkspur ferry daily, and this path would’ve been great. I haven’t ridden the new 1-2 block path on Hetherton, but going from the end of current Puerto Suello hill path to Fransico Bl W was the most dangerous part. In particular, cars exiting Hwy 101 and going right on red aren’t looking for bikes/peds, only cars come from Mission.

    At least Irwin and Du Bois were quiet and Anderson has a wide bike lane (but I always hated going N/B exiting the Cal Park path and crossing Anderson). Still the path would be great (and I am sure that area is much busier during the work day than the early mornings & early evenings that I was commuting through there).

  • CamBam415

    Going back to yesterday’s article on the MUNI & the SB Party… the Examiner said “The E-Embarcadero historic streetcar line will stop service altogether.”
    http://www.sfexaminer.com/sfmta-could-pay-super-bowl-transit-costs/

    Any idea why the E would be shuttered? It doesn’t use Steuart St/Market St and I’d think it would be a critical connector to move fans from the Wharf Hotels to Cal Train to get to the game?

  • murphstahoe

    “Once the train is in and goes to Larkspur, you won’t need to ride your bike to the Ferry anymore!”

    – Farhad Mansourian

  • murphstahoe

    also – the Puerto Suello Hill path – that’s the one from downtown SR over the hill and down, along the freeway (there are 2 said paths between SR and Novato).

    When I first did the ride North from the Ferry I found the path, off of Mission. Then when I was riding South, I kept waiting to end up on that path, but suddenly I was in downtown San Rafael never having seen that path. How do I access that path going SB? I went up over some hill on the roadway instead and descended right into town….

  • CamBam415

    Yeah, the Puerto Suello hill path goes from Mission Ave to Merrydale Ave. To access the path going south, you either need to ride south on Merrydale from N. San Pedro (there is a short steep hill up Merrydale, but minimal car traffic) or you can go up Los Ranchitos (guessing this is the way you went?) and there is an entry point across the street when Los Ranchitos turns to parallel Hwy 101 at the top of the hill. There is more car traffic on Los Ranchitos, but there are also class II bike lanes and the hill isn’t as steep as Merrydale.

    In terms of bike paths between SR and Novato, I think there are three? Puerto Suello Hill, Marinwood to Novato (Entrada Dr) and a short one from Ignacio to Novato Bl. That said, I never commuted north. Usually, I am out that way recreationally and take Lucas Valley to Nicasio to Pt Reyes/Petaluma Rd., which is way nicer than fighting traffic in Novato, but probably much slower if you are commuting.

  • david vartanoff

    unless of course I want it for further use after using the ferry.

  • murphstahoe

    “Bring your bike on the train, simpleton”

    – Farhad Mansourian

    “Umm.. trains full of bikes, sorry, wait for the next train”

    – Farhad’s conductor

  • david vartanoff

    been to that movie in Sacto a couple times.

  • njudah

    the SFMTA bureaucracy doesn’t support the E or F lines and is trying to undermine both whenever possible.

  • njudah

    if Nextbus was an inaccurate as the article claimed, I’d never use it. Sure there are times when you get a ghost train or there’s a delay whereby you see a bus coming and then it’s delayed, but that’s often due to weird things happening on certain lines that have bottlenecks. I use Routesy almost daily and use the bus shelter readouts, and most of the time, it seems to be working . I guess since I single sourced that, my mileage may vary.

  • murphstahoe

    I use Routesy almost daily

    I see what you did there.

  • njudah

    What the f did I “do”?

    Please elaborate your “comments” are so witty and clever!

  • joechoj

    So, are there any checks on schedule accuracy by NextBus?
    It sounds like it’s just a countdown of published schedules. If that’s all it is, I’d love to know so I never bother using it again.
    I’d assumed it was doing some amount of bus tracking.

  • Jeffrey Baker

    In most instances NextBus is modelling the arrival time based on the position of the bus, the time of day, day of the week, and such things. But near the beginning of the line they don’t have the bus position, because it’s not operating. That’s why it doesn’t work near the terminal.

  • joechoj

    How near is ‘near’? This was 1.6 miles from the start of the line.

    Anyway, your understanding is they do verify bus position? How, with a GPS tracker?

  • > They have no idea if there’s a vehicle coming or not.

    Do you even NextBus bro? Read the asterisk footnote.