[MTAdev] Station Entrance Locations

49 views
Skip to first unread message

Aaron Donovan

unread,
Jul 2, 2010, 10:02:09 AM7/2/10
to mtadeveloperresources
An early happy Fourth of July to all!

As requested, the MTA has posted latitude and longitude coordinates
for all individual subway station entrances.

http://mta.info/developers/sbwy_entrance.html

-Aaron

Steven Romalewski

unread,
Jul 2, 2010, 11:46:53 AM7/2/10
to mtadeveloperresources
Aaron,

Thanks for the heads up. I look forward to integrating the data.

I noticed some things in the data, however, that I had questions/
suggestions about:

- the StationEntrances.csv file doesn't include the "stop_id" values
from the GTFS data. In order to relate the entrances to the geocoded
stop locations from GTFS we'll need the IDs. Without the unique IDs
we'll have to rely on the station name, which is not only imprecise
but it appears that the names in both files are different.

- the station entrance file includes the "Division". Can this be
included in the GTFS data as well? Since NYCT organizes information
by division, it might be helpful for developers and the general public
to be able to organize and analyze data in this way too.

- the "Line" field refers to the "line segment of the station". How
does this relate to either the routes, trips, and/or shapes data in
GTFS.

- the lat/lon values have implied decimal places. Obviously it's easy
enough for us to add the decimal places in order to actually display
the entrances on a map, but it would save us a step if MTA included
the decimals in the first place :)

Thanks very much.

Adam Ernst

unread,
Jul 2, 2010, 11:56:07 AM7/2/10
to mtadevelop...@googlegroups.com
> Thanks for the heads up. I look forward to integrating the data.

I'm another developer and I'm also looking forward to integrating this data. (Of course this means my replies below don't speak for the MTA.)

> - the StationEntrances.csv file doesn't include the "stop_id" values
> from the GTFS data. In order to relate the entrances to the geocoded
> stop locations from GTFS we'll need the IDs. Without the unique IDs
> we'll have to rely on the station name, which is not only imprecise
> but it appears that the names in both files are different.

Since they're two separate databases, trying to include GTFS stop_id values could get thorny if those ever change--as they well might. It should be trivial to match by using the lines that service each stop, as well as geographical location (latitude/longitude, except for the occasional case when stops overlap).

> - the station entrance file includes the "Division". Can this be
> included in the GTFS data as well? Since NYCT organizes information
> by division, it might be helpful for developers and the general public
> to be able to organize and analyze data in this way too.

This has more to do with routes than stops. Seems useful, but also easy to compile on your own since there are only around 20 routes.

> - the "Line" field refers to the "line segment of the station". How
> does this relate to either the routes, trips, and/or shapes data in
> GTFS.

It has more to do with the conceptual naming of tracks than anything in GTFS data. For example, the 2/3 is primarily the "7th Avenue Express", but also runs on segments called Eastern Parkway, Clark Street, New Lots, etc. etc. As far as I know this data is not encapsulated in GTFS.

> - the lat/lon values have implied decimal places. Obviously it's easy
> enough for us to add the decimal places in order to actually display
> the entrances on a map, but it would save us a step if MTA included
> the decimals in the first place :)

Agreed on this point.

Adam

Steven Romalewski

unread,
Jul 2, 2010, 12:54:27 PM7/2/10
to mtadeveloperresources
Thanks Adam. True, GTFS and the entrances are two separate databases
for now (and perhaps it'd be easiest if they remained separate). But
it seems to me (and, I think, to others) that it'd be helpful to have
them integrated. Here's a thread from the GTFS Change group along
those lines (http://groups.google.com/group/gtfs-changes/msg/
0b0ddb33d4768891) and OpenPlans had already scraped the entrance
locations from NYC DoITT and included them in a GTFS-like format
(http://openplans.org/developers/data/).

When you say it'd be "trivial" to match the entrance.csv file with the
GTFS stops, I immediately assume it'd be anything but :) (and even you
acknowledge there'll be exceptions). So hopefully we can avoid errors
and problems if MTA creates the linkages internally and provides the
information in an integrated way in the future.

David Turner

unread,
Jul 2, 2010, 2:08:43 PM7/2/10
to mtadevelop...@googlegroups.com
On Fri, 2010-07-02 at 11:56 -0400, Adam Ernst wrote:
> > Thanks for the heads up. I look forward to integrating the data.
>
> I'm another developer and I'm also looking forward to integrating this data. (Of course this means my replies below don't speak for the MTA.)
>
> > - the StationEntrances.csv file doesn't include the "stop_id" values
> > from the GTFS data. In order to relate the entrances to the geocoded
> > stop locations from GTFS we'll need the IDs. Without the unique IDs
> > we'll have to rely on the station name, which is not only imprecise
> > but it appears that the names in both files are different.
>
> Since they're two separate databases, trying to include GTFS stop_id values could get thorny if those ever change--as they well might. It should be trivial to match by using the lines that service each stop, as well as geographical location (latitude/longitude, except for the occasional case when stops overlap).

GTFS stop_ids for the subway don't change, because they're some sort of
internal ID that the MTA's systems use. You'll note that some Q stops
are listed as D## -- that's because what's now the Q used to be the D
(roughly). If the stop_ids haven't changed in that long, they're not
likely to change in the future.

I'm hoping that one day this data will be part of the GTFS, and I'm
working on a proposal to make that happen. See this thread:

https://groups.google.com/group/gtfs-changes/browse_thread/thread/241013e6216a0256?hl=en

I know that this is a complicated proposal, but it is designed so that
only the relevant pieces need be implemented. Since it's not an
official part of GTFS yet, I can understand why the MTA wouldn't want to
jump in yet. But if others support this goal, supporting (or improving)
the proposal would be a good way to do that.

Kaufman, Sarah

unread,
Jul 6, 2010, 9:21:10 AM7/6/10
to mtadevelop...@googlegroups.com
Hi all,
Thanks for your input on the station entrance locations data set. I'm going to bring your comments back to the group here in charge of generating the data, and want to make sure I have all requests in hand. So far, here's what I have:

1. Add decimals to the lat/long values.
2. Coordinate the stop names with the GTFS stop names (the difference resulted from the data sets originating in different departments, but shouldn't be too difficult to fix)
3. Integrate the stop_id values into this data set.

Anything else? Please let me know and I'll get your requests in.

Also, it's in the description file, but I wanted to stress that the lines refer to rush hour only, so the secondary and tertiary lines that go through a given station may not apply during off-peak hours.


Thanks,
Sarah

Joe Hughes

unread,
Jul 6, 2010, 9:29:36 AM7/6/10
to mtadevelop...@googlegroups.com
Thanks, Sarah.

One thing that's not on your list is working with the GTFS group to express this information in the proposed GTFS format for entrance information that David Turner mentioned.  This would help move this proposal forward and ensure that future GTFS entrance info specs work well with the MTA's data.

Cheers,
Joe

Steven Romalewski

unread,
Jul 6, 2010, 10:54:19 AM7/6/10
to mtadeveloperresources
Thanks Sarah.

I noticed two other specific data issues.

One is that one of the records has a longitude value that's positive
(which means it'll be placed on the other side of the globe). It's
one of the 34th Street entrances (but the Division and Line values are
blank, so I can't identify it further without an ID number, except
that the three routes listed for this record are the ACE).

The other is that two of the records have blank values for Division
and Line: the one described above, and the record at latitude
-74001209, longitude 40731086 with routes ABCDEFM.

Hope that helps.

Steve
> https://groups.google.com/group/gtfs-changes/browse_thread/thread/241...
>
> I know that this is a complicated proposal, but it is designed so that
> only the relevant pieces need be implemented.  Since it's not an
> official part of GTFS yet, I can understand why the MTA wouldn't want to
> jump in yet.  But if others support this goal, supporting (or improving)
> the proposal would be a good way to do that.- Hide quoted text -
>
> - Show quoted text -

Samuel Wong

unread,
Jul 6, 2010, 11:03:39 AM7/6/10
to mtadevelop...@googlegroups.com
Based on my knowledge, but I could be wrong. 

The second one should be IND and the 8th Avenue/6th Avenue.  West 4th Street is the cross-section between the IND (ACE) /BMT (BDF) lines.  I can be wrong because the West 4th Street Tower is situated there and can be the dividing line.

Penn Sta/34th Street (ACE) should be 8th Avenue and IND.
--
Samuel Wong
OpenPlans | 148 Lafayette St, Floor 12, New York, NY 10013
http://openplans.org | w/f (917) 388-9042 | m: (646) 484-9142 | @samjwong
swong.o...@gmail.com
Please consider the environment before printing this message.

Government is a trust, and the officers of the government are trustees. And both the trust and the trustees are created for the benefit of the people. -Henry Clay

Michael B. Justice

unread,
Jul 6, 2010, 11:11:05 AM7/6/10
to mtadevelop...@googlegroups.com
West 4th was never a stop for BMT. Both levels serve ex-City (IND) lines. Given the recent re-routing of the M to Forest Hills via 63rd and 6th, the lines are correct.

MJ
From: Samuel Wong <samwo...@gmail.com>
Date: Tue, 6 Jul 2010 11:03:39 -0400

Samuel Wong

unread,
Jul 6, 2010, 11:15:58 AM7/6/10
to mtadevelop...@googlegroups.com
Michael,

Thanks for confirming.  West 4th was IND before using the storage yard north of the sta prior to having the 6th Avenue Express.

John L

unread,
Jul 7, 2010, 8:24:49 AM7/7/10
to mtadevelop...@googlegroups.com

Hi Sarah,
Is this just the subway entrances? What I mean by this, are the Grand Central Terminal entrances included in the dataset? For instance, Vanderbilt ramp, Hyatt cross passage, 48st passages, Helmsley passages.

Same question for my friends at LIRR for Penn Station and Atlantic Terminal.

On Jul 6, 2010 11:16 AM, "Samuel Wong" <samwo...@gmail.com> wrote:

Michael,

Thanks for confirming.  West 4th was IND before using the storage yard north of the sta prior to having the 6th Avenue Express.



On Tue, Jul 6, 2010 at 11:11 AM, Michael B. Justice <michael....@gmail.com> wrote:
>

> West ...

Kaufman, Sarah

unread,
Jul 7, 2010, 9:39:35 AM7/7/10
to mtadevelop...@googlegroups.com

John,

The file includes any entrance dedicated to the subway – for Grand Central, that includes 16 entrances; I believe all of them come in from the street and do not travel through the MNR system elements, such as the Helmsley passages.

 

Sarah

 

 

 

From: mtadevelop...@googlegroups.com [mailto:mtadevelop...@googlegroups.com] On Behalf Of John L
Sent: Wednesday, July 07, 2010 8:25 AM
To: mtadevelop...@googlegroups.com
Subject: Re: [MTAdev] Re: Station Entrance Locations

 

Hi Sarah,

Reply all
Reply to author
Forward
0 new messages