Unplanned Subway Service Changes

[itmaybeokay]

I could certainly pull off something like this on a private web server masquerading as mta.info with the help of a private DNS server, so it is possible that this is real.

 

However, it is also my expert opinion that this kind of work require a lot of man-hours:

• interface: web interface and assets
• data: directed graph structure representing all the elements of the subway system and possibly bus system
• executable code: alternative route finder algorithm

Most of the work will be concentrated on the latter two items with the data consuming the most amount of labor. It’s likely that the MTA itself hasn’t even developed this.

 

If he hasn’t actually developed it yet, he’s going to take a year to “debug” these buggies, assuming he is working alone.

 

@Javier: fix ’em quick and show us.

 

Yeah - the fact that it says "mta.info" in the address bar is a big ol redflag for me. Sure, yes - it's fairly straightforward to, say, edit your /etc/hosts to point mta.info to localhost. 

 

But that, without a doubt, would be the dumbest way to access your development platform. 

 

Also, assuming the descriptions of reroutes were automatically generated, you'd have fixed ways of describing each type of service change. The fact that the page displays "service operates in two sections" "is split into two sections", inconsistently applies capitalization, boldface and italicisation kinda screams human-generated document. If this were generated by a computer, the natural-language processing would seem to rival that of IBM's Watson. 

 

 

But, I'll give the benefit of the doubt if you answer me these questions three: 

What language is this written in? How is the representation of the possible reroutes stored? What path-finding library are you using? 

Edited May 17, 2016 by itmaybeokay

[Javier]

Must I remind you, this is a beta. There may be multiple codes that all have the same output, hence why it has a different service change output. Plus, all stations have a precoded example service template. But I'll answer your questions:

 

C#

 

The way how they are stored is that each station has a pre-coded example of the possible service changes that can occur. However, I need to manually input all the track diagrams into code form so the servers/computers know how to reroute the congestion, although I've only imputed it in for some stations in Manhattan, hence why the computer/server bugged out with the 42nd St shuttle and the reroute.

 

And the path finding library I'm using is Java, as it is the quickest. I may change it to Python later, but I'll see how it goes.

[itmaybeokay]

Must I remind you, this is a beta. There may be multiple codes that all have the same output, hence why it has a different service change output. Plus, all stations have a precoded example service template. But I'll answer your questions:

 

C#

 

The way how they are stored is that each station has a pre-coded example of the possible service changes that can occur. However, I need to manually input all the track diagrams into code form so the servers/computers know how to reroute the congestion, although I've only imputed it in for some stations in Manhattan, hence why the computer/server bugged out with the 42nd St shuttle and the reroute.

 

And the path finding library I'm using is Java, as it is the quickest. I may change it to Python later, but I'll see how it goes.

 

Your C# web application, which I'm assuming is running on an ASP.NET platform - is calling a Java Pathfinding library (unnamed) , presumably by loading a virtual machine for java... because that is the fastest?

 

IDK man, share some code and I'll happily retract my doubts and praise your ingenuity but none of this checks out to me.

 

The most widely used pathfinding library is A* which is easily implementable directly in C# and source code of such is publicly available.  

 

As for each station having pre-coded examples of possible changes, wouldn't that result in a list of variables several thousand entries long? 

 

Wouldn't it be easier to make a representation of the trackwork as a boolean grid, and like descriptors of the theoretical changes based on nodes the path-finding algo goes through ? (Via the X line between Y and Z, Suspended between X and Y, Running in two sections between A and B and X and Y, On the Local/Express track between Q and R) 

 

All you have to do is change one of the tiles on the grid from TRUE to FALSE and the pathfinder has to find another way. 

 

This way has it's drawbacks but it's probably the most straightforward and versatile way of putting it together. 

 

if you're spoofing a goof here tell us.

if you've realllllly got this running, even as buggy beta town - put the code on github because now I'm interested and would honestly love to work on it. 

[Javier]

Wow...I could've used a boolean...and yea this has a thousand entries for every service change I can think of, but since you all want to test it so bad, I'll make it into a stand alone program and then release it with source code when I get the chance.

[itmaybeokay]

Wow...I could've used a boolean...and yea this has a thousand entries for every service change I can think of, but since you all want to test it so bad, I'll make it into a stand alone program and then release it with source code when I get the chance.

 

:-D that would be awesome. 

[Dj Hammers]

Wow...I could've used a boolean...and yea this has a thousand entries for every service change I can think of, but since you all want to test it so bad, I'll make it into a stand alone program and then release it with source code when I get the chance.

in the mean time, can you please upload the code to GitHub?

[Lance]

I'm surprised you're all entertaining the idea this is anything more than a Photoshop job or something similar. This isn't the first time this has happened, now is it? The whole thing is amusing though.

[P3F]

I'm surprised you're all entertaining the idea this is anything more than a Photoshop job or something similar. This isn't the first time this has happened, now is it? The whole thing is amusing though.

Frankly, I also doubt that this code exists. But this post was simply rude. I would expect more professionalism from a "Senior Moderator".

[Around the Horn]

Frankly, I also doubt that this code exists. But this post was simply rude. I would expect more professionalism from a "Senior Moderator".

He has a point. Javier, god bless him, has done this before. Remember the whole December 2016 timetable debacle?

 

Till I see this app, I'm treating it as an elaborate photoshop job.

[P3F]

He has a point. Javier, god bless him, has done this before. Remember the whole December 2016 timetable debacle?

 

Till I see this app, I'm treating it as an elaborate photoshop job.

That is true. However it is still my opinion that Lance's post was inappropriate for a staff member of a forum.

Edited May 18, 2016 by P3F

[Javier]

He has a point. Javier, god bless him, has done this before. Remember the whole December 2016 timetable debacle?

 

Till I see this app, I'm treating it as an elaborate photoshop job.

That was an early April fools joke .

Guys, I will show you the code can you just be patient. I don't have to show/give it to you guys you know...

[Daniel The Cool]

That was an early April fools joke .

You know what I don't believe that because thats not the first time you did that. Whatever happened to those Trains you said that went to Kings Highway? 

[CenSin]

That was an early April fools joke .

Guys, I will show you the code can you just be patient. I don't have to show/give it to you guys you know...

Have you heard of a story called “The Boy Who Cried Wolf?”

 

There’s been a long history of such accusations and not only on April Fools Day.

 

You are a psychotic liar

Tell you what, show an ACTUAL source that can be found on the web (and not some random person that is somehow connected to the ; whether or not you are lying it could be proven false without some form of document) and then we'll talk.

Stop lying. The is not running to Brooklyn Bridge today due to a G.O. SMH

There was another time early on when you said something. Although I can’t find that thread, I remember distinctly that your response to a comment about the veracity of your claim was “are you calling me a liar?” That was probably way back in the early 201Xs.

 

I want to believe, which is why I did not say anything this time, but when you say something there is often plausible evidence that supports doubt.

[Javier]

Ok and I get it, I had that problem before. Not now. Let's move on from this discussion and get back on-topic. When it's ready I'll post it. End of discussion.

[Lance]

Agreed.

 

Service Change  Posted: 05/19/2016  9:05AM                    

Due to a train with mechanical problems on Manhattan Bridge, southbound  and trains are running on the  line from Canal St to DeKalb Av.

Expect delays in southbound train service.

Allow additional travel time.

Edited May 19, 2016 by Lance

[CenSin]

Agreed.

Northbound trains went local around the same time. The platforms were crowded. I got on at Neck Road after waiting 10 minutes for a . I checked behind the train every few stops and there were no trains behind us.

[Javier]

Does anyone know what happened at Canal St on the today? I saw a R62A and the last car was completely dead.

[Joel Up Front]

"How you want your B Division?"

"Completely FUBAR, fam"

 

"Say no more."

 

 

Due to a rail condition at Cortlandt St, the following service changes are in effect:

Some southbound trains are running on the line from Canal St to Dekalb Av.

Some southbound trains are running on the line from 36 St (Queens) to West 4 St-Washington Sq, then over the line from West 4 St-Washington Sq to DeKalb Av.

Expect delays in train service in both directions.

Allow additional travel time.

[P3F]

Took the south from Whitehall at around 3:45 and it was fine.

 

Northbound trains were running via the bridge, though.

Edited May 23, 2016 by P3F

[NBTA]

Well then...

 Service Change  Posted: 05/25/2016  2:17PM
 

Due to a rail condition at Central Park North-110 St, the following service changes are in effect:

Both directions, there is no train service between Harlem-148 St and 96 St.

Southbound trains are running on the  line from 149 St-Grand Concourse to Nevins St.

Passengers may use the Bx19 bus which runs along 145 St for service between Malcolm X Blvd and Broadway.

Allow additional travel time.

[TrainFanInfinity]

Took the south from Whitehall at around 3:45 and it was fine.

 

Northbound trains were running via the bridge, though.

 

Yeah there was some "FDNY activity/investigation" at Canal Street/City Hall.

 

Sooo about that rail condition at Myrtle-Bway this morning, heard an got rerouted to Jamaica Center for some reason.

[BDNQ2345]

 

here the via line to Broad St

Edited May 25, 2016 by BDNQ2345

[MTA Bus]

Does anyone know what happened at Canal St on the today? I saw a R62A and the last car was completely dead.

12-9

[MysteriousBtrain]

Due to a report of an unauthorized person on tracks at 42 St-Bryant Pk, the following service changes are in effect:

There is no  train service from 145 St to Brighton Beach in both directions.

Southbound  trains are running on the  line from 59 St-Columbus Circle to W 4 St-Washington Sq.

Southbound  trains are running on the  line from Jackson Hts-Roosevelt Av to W 4 St-Washington Sq.

There is no  train service from Delancey St-Essex St to Forest Hills-71 Av in both directions.

Police are investigating.

Allow additional travel time.

[Joel Up Front]

What is it with the and ?

 

 

 

Due to hanging wires at Bowling Green, the following service changes are in effect:

Northbound  and trains are running with delays.

Some northbound  trains are running on the line from Nevins St to 149 St-Grand Concourse.

Allow additional travel time.