6 ms·
Building Waze for the Boston subway: my first adventure in civic hacking
- chiph 12y agoNot actually Waze - title would be better along the lines of: "Building a Waze-like for the Boston subway"
- ryanthejuggler 12y agoI think the title was written in the spirit of "It's Uber for <X>". Agreed that it could be clearer though.
- gklitt 12y agoYeah, this was the intention. As cliche as "X for Y" has become these days, I can see why it's such a popular format. Most people instantly get what MBTA Ninja is, based on a four word tagline.
- mayneack 12y agoAlthough, Boston is partnering with Waze for their non-subway congestion problems: http://www.cityofboston.gov/news/default.aspx?id=18994 http://www.cityofboston.gov/news/default.aspx?id=18994
- sethbannon 12y ago"In a world where political systems are increasingly gridlocked every day, and much of Silicon Valley is focused on peddling ads, the civic innovation and open data movements are a bright and optimistic exception to the zeitgeist." Amen to this! I'm excited by the trend of more hackers realizing that they have the skills to fix a lot of what is broken about politics and government.
- aepearson 12y agoAmen, x2!
- hugenerd 12y agoThis seems similar to is metro broke? https://itunes.apple.com/us/app/is-metro-broke/id849428357?mt=8 https://itunes.apple.com/us/app/is-metro-broke/id849428357?m...
- untog 12y agoThis could be very interesting, and could even sit in the background and track user's movements without needing an explicit action to report problems. Unless you're in New York where everything is underground of course. Sadly.
- mayneack 12y agoMost of the MBTA is underground too. However, at least the green and red lines have cell reception underground.
- bwanab 12y agoThe Orange and Blue lines have it also. As far as I know the whole system is wired underground.
- mayneack 12y agoyeah, figured as much, but didn't want to make claims about lines I hadn't used enough to know for sure.
- antsar 12y agoThere may be reception, but is it possible to get a GPS fix? Without one, it may be hard to make automated assumptions about congestion / travel speed.
- what_ever 12y agoUse wifi information for geolocation since most of the Subway stations in NYC have wifi now?
- manacit 12y agoI would hardly say that "most" stations have WiFI in NYC. There are 421 total, and the MTA is targeting service in 278 of them by 2017. The list of stations that currently have service are as follows: http://www.scribd.com/doc/243224749/MTA-stations-that-have-WiFi-Phase-I-Phase-II http://www.scribd.com/doc/243224749/MTA-stations-that-have-W...
- chrisBob 12y agoEastbound, Northbound, Southbound, Westbound? I am not sure the authors have ever been to Boston. Here there is only Inbound and Outbound. I went to www.mbta.ninja and then spent about 30 seconds trying to figure out if any of the alerts are related to me.
- _august 12y agoInbound/Outbound changes at Park St, correct? So South/North/East/West seems to be more helpful here. Of course, they could ask you to input a station first, then display the connecting lines as inbound/outbound.
- sisk 12y agoThe four stations where the directions change are the square made up of Park Street, State Street, DTX, and Government Center (or Haymarket since Gov Center is down? Not sure, haven't been to Haymarket in at least two years). Directionality (for new folks) or terminus (for more seasoned riders) are infinitely better than Inbound/Outbound.
- swalsh 12y agoSaying "To Alewife" or "To Ashmont" is as much information as anyone really needs (if you're looking at the redline for instance)
- ethomson 12y agoNo, there is only the destination of the train, since an Alewife bound train is "inbound" or "outbound" depending on which station you're at. If you want to talk about the Red Line delays going toward Alewife, then inbound or outbound is uselessly ambiguous.
- gklitt 12y agoI actually commute on the T every day. As _august pointed out, inbound/outbound switches at Park St, so it's not the best way to identify alerts going a single direction throughout the line. We are definitely trying to make the line selection easier to use, so if you have ideas it would be great if you could make a Github issue! https://github.com/davidlago/mbta-ninja https://github.com/davidlago/mbta-ninja
- pimlottc 12y agoI like the confirmation model. And mbta.ninja is a cool domain name, though why besmirch it with a needless www?
- cllns 12y agoI had a lackluster interview experience at the company the author works for and mentions. After completing a coding challenge, I got a generic: "After deliberating, we have decided not to proceed with your application." I asked for more specific feedback, from two different people, and got no response.
- swanson 12y agoThis seems completely unrelated to the article and doesn't really add anything to the conversation.
- cllns 12y agoI mentioned it because the company itself was plugged in the post. Also there's a bit of irony involved, where they're extolling the virtues of data, but failing to provide proper interview feedback.
- crazypyro 12y agoThat's pretty much standard operating procedure. You don't want to open the company up to a legal challenge as an interviewer...
- kingnothing 12y agoThat's generally the only feedback you'll get from any company.
- cllns 12y agoThat's not true. I've gotten better, more detailed feedback from other companies. Things like: your approach for this coding challenge was X when you should have been doing Y.
- bitwize 12y agoSounds a lot better than going into the station blind, thinking you'll make it for morning team meeting, only to hear the voice of GLaDOS tell you how shitty your commute is actually going to be.
- this_user 12y agoThis idea has a lot more potential. Once you have the data, you can obviously give the individual user the best alternative route if there is delay on the desired one. You can also use this proactively, similar to what "Google Now" does, to alert commuters in the morning how much longer their trip might take. You could even offer functionality to automatically adjust the phone's alarm clock. But you can take this even further than individual users and dynamically re-route passenger flows if enough of them are using the service. Say there is an interruption on a main commuter route, but several alternatives exist. Most people are now likely to choose the second best route. Since that route is unlikely to be able to handle the additional passengers, it will probably be jammed very quickly. However, using the RT passenger and situation data, it would be possible to find the globally optimal solution for load balancing the passenger flow. You can then provide the passengers via smartphone with individualised information regarding which route to take to implement this routing. I think this has a lot of potential for taking public transport to the next level by utilising existing capacities to the fullest extent and providing benefits to all passengers.