Talk:List of London Underground stations

From OpenStreetMap Wiki
Jump to navigation Jump to search

Use aerial imagery instead of GPS (?)

It's no longer necessary to gather GPS readings from tube stations. We just have to pinpoint them on Yahoo! Aerial Imagery (see that page for datails) ...or am I missing something? -- Harry Wood 15:29, 26 January 2007 (UTC)

This list is Public Domain, not CC, so you can't copy from OSM data, you have to have collected the points yourself. My understanding is the Yahoo agreement is for data to be extrapolated for OSM under a CC license, which would prevent you from posting coords here... it's entirely possible I'm wrong about that though. Randomjunk 20:17, 26 Jan 2007
Ah yeah. What a pain. I just added a node pinpointing the exact location of Camden Town tube station based on the Aerial imagery. I was somewhat surprised that nobody had already done so. I'm happy to release this contribution as public domain, if I am allowed to do so, but as you say, strictly that might be in breach of our agreement with Yahoo! I dont know the exact legalities of the agreement
...but really that's getting a bit silly now isn't it? I mean you could tweak the lat/long digits a little and call it public domain. Who's going to know? :-) Of course if we extend that argument to several nodes, and interlinking segments, then hey presto! we have a public domain streetmap.... hmmm. maybe I shouldn't give people ideas. Bah! Annoying legalites -- Harry Wood 18:11, 12 February 2007 (UTC)

Tube Map

Once we've got a fair number, we can look into overlaying them on-top of some OSM data, to produce a free tube map.

I've also been adding the tube stations I survey as nodes in OSM, and joining them up when I have traces for the route, which should help. Gagravarr 14:41, 27 September 2006 (BST)

worth writing down node numbers in a list somewhere?
We could make it another column on the table, if people think it'll be useful. FYI, I've been tagging mine with railway=station and tube=District, Circle, Piccadilly, so they're easy to search for Gagravarr 15:22, 27 September 2006 (BST)

What would be really neat is to use that physical model technique that's used for graphing topic maps and social networks and things, to create a topological (is that the word?) map that hopefully turns out something like the official map, but being algorithmically generated from our geodata, can't possibly be considered a derivative work by LU. I think the process goes something like:

  • Start with all station nodes in their true geographical locations
  • Consider each station in turn and update its position, treating any two stations that are joined by a line as exerting an attractive force, and any other two stations as exerting a repulsive force (possibly proportional to their geographical distance?)
  • Repeat for 'n' iterations, until the system becomes stable

Mattwestcott 20:39, 23 January 2007 (UTC)

That reminds me of this GraphLayout demo applet. Fun to play with! -- Harry Wood 15:26, 26 January 2007 (UTC)


My dates all just say (2006) as I've been collecting locations in my eTrex for a while but have no real idea when. --Parsingphase 14:04, 9 December 2006 (UTC)

You've listed Elephant and Castle entrance - but it has two entrances! How should I flag that up? Secretlondon 13:31, 8 March 2007 (UTC)

Error checking

One should probably provide links to a copyrighted map so that obvious errors can be spotted and data removed. West Hampstead for example has been located in Putney... Ed gps 16:36, 30 January 2007 (UTC)

We could just track the OSM node ID in the list (since you're free to put this data into OSM, just not find places in OSM and list them here). That'd allow us to be sure all the stations listed have been put into OSM, and from the node ID we could display the location on the slippy map (via a web service call or similar), which people could use to check. Gagravarr 19:05, 30 January 2007 (UTC)
We could also accept multiple entries for each station to get an average. This should be closer to the actual value and also, it is what is done when mapping using GPS. If there is a small spread on the way I and many others would instinctively place the nodes and segments in the middle of the data. Accepting multiple entries amd calculating an average would mimic this. In case this approach is used, I have a value of 51.48153±0.000025, -0.01063±0.000035 for Cutty Sark taken from 3 GPS readings. Also I am happy to calculate the statistics. Ksbrowntalk 19:14, 16 April 2007 (BST)

I've just visualised this data and notice that there are a chunk of stations listed as being in the Channel instead of in London. They can be identified by their latitude being what looks like ~1 degree off to the south (~50.5 instead of ~51.5) --Dankarran 15:28, 15 June 2007 (BST)

I think I've made a number of typos of this sort. Can you give a list of dodgy ones, and I'll retype from my GPS (which still has the points saved)? Morwen 15:31, 15 June 2007 (BST)
The ones that look to be obviously off are: Surrey Quays, Wapping, Westminster, Whitechapel. Dankarran 22:11, 16 June 2007 (BST)

Data for Gloucester Road is incorrect (it matches South Kensington). GAnders 15:58, 21 March 2008 (UTC)

Tube data for Wikimedia Commons


We collected a little bit of data for our Geographic maps of the London Underground for Wikimedia Commons. It's PD, and you're welcome to it.


Jdforrester 01:54, 4 March 2007 (UTC) (James F., Wikimedia)

I just found (and re-assembled) the following discussion and moved it here because it was on the Talk:Free The Postcode page (where tube station efforts started out). Anyway the following looks like it goes some way to explaining why we're not using the wikimedia commons data -- Harry Wood 10:33, 18 April 2007 (BST)

There is a complete(?) list of tube station locations collected with GPS for the Wikimedia Commons geographic tube maps: -- TomC - 10:12, 13 February 2006

Yes, but nobody knows where it came from. At least, as far as I've found.
And, looking at the data, doesn't look like GPS. - Steve 10:54, 13 February 2006
The wikimedia commons page states that it's sourced 'solely from GPS'
Just because it's all 4dp doesn't mean anything - nobody said it was raw GPS, just that it was sourced from GPS. Nevertheless, since we now (now you've asked) have reason to believe that Google Maps was also involved in the collection of the data (which obviously makes OSM cry) then, erm... "as you were, soldier!" TomC 15:26, 14 Feb 2006 (GMT)

As I understand it, from talking to a friend who's involved heavily in Wikimedia and understands this sort of thing (David Gerard), the Wikimedia Commons data is public domain, because the data comes from a NASA image, which is public domain. Regardless of whether that's right or not, I think the existence of multiple independent datasets is a Good Thing; plus, I like the idea of a project that involves people leaving their computers and going out into the actual world - that's why I'm adding to this page rather than just going "oh, problem solved, no need to do anything more". Kake 10:24, 27 April 2007 (BST)

Hang on, he's saying that NASA provide public domain imagery at a high enough resolution to identify tube stations? Shouldn't we, kinda, know about this? Mattwestcott 13:11, 7 May 2007 (BST)

Another project people here might be interested in.

On the Randomness Guide to London, we've been collecting free geodata for things like pubs and restaurants as well; if you like collecting GPS data then you might like to wander over there and give us a hand. --Kake 11:27, 28 April 2007 (BST)

next project

Ok, now for all the rest of the railway stations? Morwen 12:10, 17 August 2007 (BST)

Rest of railway stations - Say everything within the M25? (Nice clearly defined goal) Firefishy
Everything within the Travelcard zones? Morwen 15:15, 17 August 2007 (BST)
Sounds good to me, lets do it. Firefishy 15:36, 17 August 2007 (BST)
Is there a page for this? I can't find it. Kake 08:50, 27 November 2007 (UTC)
Just add them to this list for now, or start a new table. Morwen 09:23, 27 November 2007 (UTC)

Add to the map

Can we make sure all these tube stations are actually added as nodes to the map data? That might seem obvious, but they're not all on there. How can we check easily? Render a map of only tubes I guess -- Harry Wood 13:05, 18 August 2007 (BST)

tube clean up

  • Link Data Eg: 1 -> 2 -> 3 -> 4, 3 -> 5 etc)
  • Adding missing Line details Eg: Ealing Broadway on Central, District

Oxford Circus Station

If I am not mistaken, Oxford Circus Tube Station has disappeared ? I cant believe this was an oversight so I guess someone must have deleted it? Can someone resurrect the station from the history? --Petz 14:33, 18 February 2008 (UTC)

Done. Added again, allow a few days for the renders to catch up. Firefishy 20:10, 18 February 2008 (UTC)