Talk:London Cycle Network

From OpenStreetMap Wiki
Jump to navigation Jump to search

LCN route numbers in a state of flux?

Is it just me, or are the London Cycle Network route numbers in a permanent state of flux?

The cycle track that passes my road is known as one of the following:

1, 6 or 85. 1 according to a map on an information board. 85 according to a London Cycle Network sign at Staples Corner, showing a primary destination of Barnet. 6 according to the LCN+ route maps.

According to the London Cycle Network web site, apparently routes were being renumbered a few years ago, and there's no obvious result of whether the renumbering decisions have been completed yet.

That said, I wouldn't trust the LCN+ web maps for directions. They've sent me along many a route that has plenty of "no cycling" signs up.

Welshie 23:59, 27 May 2008 (UTC)

Should all cycleways be tagged lcn=yes?

In London I've come across a few different kinds of cycleways that don't consistely appear tagged and I'm not certain whether they should.

  1. signed routes with numbers (e.g. [6]) or destinations (e.g. West End 2 miles)
  2. routes signed with rectangular blue bicycle signs (possibly including mandatory cycle lane or contraflow signage)
  3. routes with circular bicycle signs, big ones in intersections, small ones on bollards and such
  4. cycle lanes and tracks, bicycle signs painted on the road
  5. bicycle signs painted on the road
  6. bus lanes signed for cycles as well (guess that's all of them)
  7. signed directions to a (numbered) cycle route

I think obviously 1) is lcn=yes, but for the rest it could be argued both ways, and maybe even depends on location. My impression is that in London all ways explicitly marked for bicycles one way or another ways are part of the cycling network whereas in Finland majority of pedestrian paths are also cycle paths and you'd only tag signed routes.

--Tko 12:07, 17 July 2010 (UTC)

I was wondering about the same thing while cycling in London. To my knowledge rectangular cycling signs are "advisory" (with no provisions) whereas the round shaped ones mark paths dedicated to cyclists. Therefore I'd say the circular bicycle signs should be tagged with "lcn=yes" and the rectangular ones with "lcn=yes" with possibly "state=advisory" Thoughts? -- Tilusnet 14:50, 19 July 2010 (UTC)
My understanding is that rectangular signs are mainly for cars to highlight assigned lane usage, e.g. left lane must turn left and right must go straight on, or one way with cycle contraflow, or car lane with mandatory (or advisory) cycle lane, or common lane shared with cycles. Circular signs just say "no cars". I get the feeling shared or dedicated cyclepaths (round signs) are mostly just in places like parks and thus less likely to all be part of cycling network. OTOH cycle lanes usually feel more like part of a network getting you somewhere. --Tko 20:35, 19 July 2010 (UTC)


Tagging

Following is my understanding of how certain cycleways should be tagged. I think lane/opposite_lane/track/lcn_ref/ncn_ref are clear and the ambiguous ones are recommended routes, bus lanes and lcn=yes. --Tko 22:32, 27 July 2010 (UTC)

Description Photos
Cycles on the road with other traffic
Route recommended for cyclists on the main carriageway of a road.

bicycle=designated (based on reading [Cycleway] where none of the values apply, except the implied one)

I'm not completely certain, but these markings (sometimes they look like cycle lane markings without the line separating the lanes) also seem to be used to reinforce cycle routes between destination signposts. Shouldn't they be lcn=yes or do you need the sign with explicit lcn too?

Note: currently not rendered in OpenCycleMap [1]

UK bicycle designated.jpg UK bicycle designated example.jpg
With-flow cycle lane

cycleway=lane

Or does this even apply to advisory cycle lanes since they have no special meaning really as you can drive and park your car there freely?

UK cycleway lane track.jpg UK cycleway lane example.jpg
With-flow bus lane for cycles (and taxis)

(not tagged)

Since there is a designated, if shared, lane for cycles shouldn't this be tagged cycleway=lane?

Contra-flow cycle lane

oneway=yes

cycleway=opposite_lane

UK cycleway opposite lane.jpg UK cycleway opposite lane example.jpg
Cycle track

cycleway=track

Note: currently not rendered in OpenCycleMap [2]

UK cycleway lane track.jpg
 London/National Cycle Network
London Cycle Network routes (unnumbered)

lcn=yes

The other sign doesn't explicitly say "London Cycle Network" so it's basically just a sign with bicycle on it (just like the recommended route one) - is such route part of LCN or not?

UK lcn.jpg UK lcn2.jpg
London Cycle Network routes (numbered)

lcn_ref=*

Note: it is preferred to add the ways to the relation with lcn_ref=* tag rather than tagging the ways directly.

Sidenote: should the relation be named "38" (gets bit mixed with bus routes), "LCN 38" or "London Cycle Network Route 38" (most correct I'd think, but annoyingly verbose)

UK lcn ref small.jpg UK lcn ref.jpg
National Cycle Network routes

ncn_ref=*

UK ncn ref small.jpg UK ncn ref.jpg

renaming to shorter title

The page is currently called "WikiProject United Kingdom London Cycle Network". I've been renaming a load of UK related wiki pages which unnecessarily use the "WikiProject" prefix (see Talk:WikiProject United Kingdom#renaming the "WikiProject" pages). But this page is has an even more unnecessarily long name at the moment . I think it can be just "London Cycle Network" -- Harry Wood (talk) 10:48, 10 April 2016 (UTC)

Tagging to reflect LCN routes which have been turned into Cycle Superhighways or Quietways

Hi, not sure there's much activity on this page any more! However, I'm fairly new to London and have been finding cycling maps based on OpenStreetMap's data to be HUGELY more accurate than TfL's own ones - very grateful for all the past work that's been done. (For instance WaymarkedTrails [3] and OpenCycleMap [4] ) The things with these, though, is that in London they can get very cluttered with old LCN routes which have since been turned into Quietways or CycleSuperhighways. Examples CS1, which basically is a street-by-street upgrade of what was LCN10, likewise LCN11 and CS2. In cases like these, all the LCN signage and lane markings have been replaced by CS ones. Effectively, these sections of the LCN route are now obsolete and there's little point including them on maps for current use.

I'm wondering what the best way to address this is? Because most of these LCN routes aren't obsolete for *all* their length - usually just a portion has been upgraded. Do we need to duplicate each LCN's relation, and then globally tag one of the relations as 'obsolete'. And remove all the obsolete route segments from the current relation, and vice versa? --Feline1 (talk) 13:49, 13 June 2018 (UTC)

In such cases, I would be of the opinion that upgraded sections are still part of the old LCN route and that the LCN route is not obsolete in the upgraded sections. Even in the original LCN network, there were shared sections of numbered routes, e.g. along LCN 0. Similarly on the tube the Circle line shares with other lines.
I recently changed the network of Quietway/Central London Grid/Mini-hollands from network=lcn to network=rcn (in line with the Cycle Superhighways) as this London-wide network is really a regional network. This change helps to highlight the new network as different from both non-LCN local routes and the non-upgraded sections of LCN routes. (However, strictly speaking, I would say the London Cycle Network is also a rcn.)
Surprising in some places I have noticed new signage for LCN routes in recent months, so not all emphasis is on the new network.
On a slight tangent, "LCN+ links" aren't routes but are for administration only (i.e. strategic local links for project management purposes within individual boroughs). I think we should find a better way to tag these, as type=route isn't really correct and they shouldn't really appear as e.g. '+93' on maps.--MacLondon (talk) 01:23, 14 June 2018 (UTC)
Oh absolutely I agree that most of the LCN is not obsolete at all - at the current rate, it would be a couple of decades before it's all converted into QWs or CS's, and even most of the bits that have been upgraded still have big chunks of the underlying LCN route that remains... so we definately don't want to remove those from displaying on maps!
But it does look confusing, I think, on places like, for example, Southwark Bridge: this is clearly CS7 now, all signage and blue paint. And heading north at the north end is now Quietway 11... so labelling these on a map as 'LCN0' as well serves no sensible purpose. (It's not like the Circle and District tube lines where you still get trains from both sharing the same tracks and platforms... LCN0 here is just a historical relic). There are similar situations on the road network, where for instance part of a former A road might have been wholly upgraded to a motorway. In that case, maps don't still label the motorway as also being part of the A road - instead the A road just continues at either end.
I also agree that the maps shouldn't display all those LCN+ things willy nilly. For example, I've trying to find my best commute route to work, and seen on those maps that there's a 'cycle route' of 195+ or something across Tower Bridge... there isn't ;). It's very misleading to have a real-world cycle map which labels streets with what was just some office reference number from over a decade ago, which didn't necessarily result in any signage or infrastructure on the ground. Definately those shouldn't be tagged to imply something that doesn't exist. --Feline1 (talk) 08:45, 14 June 2018 (UTC)
As an experiment, I fiddled with the tags for the relation LCN + link 195 [5]. This rather pointlessly was getting shown on maps on top of CS3 and NCN13, and also making it look like there was a cycle route over Tower Bridge. If I now call up this area on, say, WaymarkedTrails [[6]] CS3 and NCN13 still get overlaid with LCN15, which is a bit pointless, but the western end of CS3/NCN13 is now more realistic, with the lack of cycling infrasturcture connections around Tower Hill shown properly. I'd wager some similar tag-culling could be done on many other LCN+ link relations, however I wouldn't want to do it myself because I don't know any of the streets in question, in some cases the local councils may have put infrastructure down on the ground which means they are worth showing on maps?--Feline1 (talk) 09:00, 14 June 2018 (UTC)
What I have ended up doing is creating new relations for the upgraded sections of the LCN, and tagged them as "route=historic" instead of "route=bicycle" (e.g. for bits of LCN19 now upgraded to Q13 This means they no longer clutter up maps of current cycle routes (e.g. Waymarked Trails ). However what needs doing next is creating an overall relation that combines bother the remaining and the obsolete bits of an LCN route together so they can still be displayed from start to finish, and for the obsolete relation also to be added to the LCN super-relation

How to deal with ongoing roll-out of Quietways and Cycle Superhighways

Currently there are a good few dozen or so Quietways and CS's that have relations with the tag "proposed". This don't appear on the excellent WaymarkedTrails map, or are dashed(?) on OpenCycleMap. However, most months that go by, some of these proposed sections get implemented on the ground. (For example, this most, some stretches of Quietway Q7 went live.) This is the other side of the coin to my question above about marking old LCN routes obsolete - what's the best way to change the status of a route from Proposed to operating? Do relations need duplicated & split out, so for instance you could have Quietway 7 (live) and Quietway 7 (proposed)?--Feline1 (talk) 13:59, 13 June 2018 (UTC)

Well, I have just been bold and removed the "proposed" tag from Q2, Q7, Q11, Q14... they're all on TfL's website as "completing 2018" (well, a man can hope ;) and considerable lengths of them exist and are signed and marked on the ground. The bits that aren't are usually just upgrades of exists LCN routes anyways, so people trying to follow them from a map will still have something to ride along... there are a few sections still in an unimplemented state, though. However even in this case, they tend to just be ordinary roads which it's possible to cycle along anyways with care. --Feline1 (talk) 21:41, 14 June 2018 (UTC)
As an improved solution, I have ended up taking the option of having two relations for Quietway 14, one for the open section, and one for the still 'proposed' (i.e. unsigned) section. As more streets get their signage, it's not too difficult to remove them from the 'proposed' relation and add them to the 'live' one. (The proposed relation has a "route=proposed" tag instead of a "route=bicycle" one). However for bonus points, it would be good to add both of these routes to Q14 super-relation, and then add that to the overall London Quietways relation ... I'll have to leave that to someone else as I only use the in-browswer Potlatch editor at the moment and I don't think it let's me edit superrelations in that way!--Feline1 (talk) 09:55, 8 July 2018 (UTC)
So, we have more or less gone for the solution of having two relations for routes that are in the process of rolling out, one with a "route=proposed" tag than gets gradually gobbled up by a second relation for the live route. Seems to work OK!--Feline1 (talk) 12:15, 12 August 2018 (UTC)

Local vs Regional route tags for Quietways and LCN routes

Greater London's Open Cycle Map and Waymarked Trails maps is beginning to look extremely busy, as we've been tagging tons of previously unlabelled routes. I think mapping might have more clarity (and be truer to the spirit of how tags were intended to work) if we tweaked things a bit as follows:

  • TfL Quietways that have a number (and typically span multiple boroughs, and are usually listed on TfL's website) should stay as "route=rcn"
  • However Quietways that don't have a number (but are just labelled "Q", are typically within one borough, and aren't listed on TfL's website) would perhaps be better tagged as route=lcn?

(I know it's not always as clear cut as this, and some of them do span multiple boroughs... many (but not all of them) are part of the 'Central London Grid' scheme... generally they are local routes that have been 'upgraded to Quietway standard' - nebulous though the definition of 'quietway standard is - rather than being a route from A to B.

There's a twitter account that has been building up a Google Map of Quietways using different colours for numbered and unnumbered ones, and I think it works quite nicely.

I also think there might well me merit in revising the tagging of some older London Cycle Network routes in an analogous fashion:

  • LCN routes that are numbered and signed on the ground typically round through several boroughs, and I think it would make more sense to label them as "route=rcn" (currently they are all tagged as lcn)... I wonder if this misnomer arose simply because the letters "LCN" and "lcn" are the same? ;)

There are probably hundreds of miles more of genuine "local" cycle routes in London. Often these have no particular route signage at all, they can range from anything to simply a traffic calmed street (eg 20mph speed limit, speed bumps with cycle cut-throughs), cycle lanes marked by the side of the road, shared bus lanes, contra-flow lanes in one way streets...some may have a 'cycle' marking on the carriageway. Recently we have (rightly, I feel) tagged more of these with lcn relations

The biggest grey area for me is numbered LCN routes that do not use route numbering on the carriageway or signage. Should these be tagged as rcn? I'd cautiously say they should - very often the use of numbering seems to be at the whim of the borough in question - and a route may well pass from one borough to the next, being clearly numbered in one, then un-numbered in the next (but usually with LCN route signage showing a named destination). So they are "regional routes" in that sense. Things do get greyer still though if a borough council never really implemented a proposed LCN route very well in the first place, and by now is certainly not maintaining what signage, markings or infrastructure they put in place.

Thoughts? --Feline1 (talk) 12:15, 12 August 2018 (UTC)

I wouldn't agree with the suggestion that unnumbered routes are in a different network to numbered routes. The tagging of routes is route=bicycle, not route=rcn. The addition of network=rcn is based on the network it is part of, and that is independent of the length of the route or how many boroughs it passes through or whether it is numbered; there are a few NCN routes in London (NCN 162/208/425) that are basically local routes, yet they are part of the national cycle network, so they therefore have to be tagged with network=ncn. Similarly any short unnumbered 'Q route' is on the exact same network as Quietway 1.
The problem with the old and new London cycle network is that it is a grid, with only some of it made up of numbered routes. In many cases especially in central London, the unnumbered Q sections aren't really routes per se; in practice, all ways from all these sections could be gathered into a single "route" relation. However, as routes are meant to be continuous, in order to map the network it works to map these sections separately like we've done using type=route.
With regard to renderers such as Waymarked Trails becoming "busy", that will inevitably happen in London with sometimes a number of cycle routes sharing the same sections of road. In any case "tagging for the renderer" is frowned upon on OSM. Basically, while a renderer could potentially be designed to use different colours for numbered and for unnumbered routes on the same network, tagging them on OSM as different network=* in order to achieve the same effect is wrong.
With LCN routes, perhaps we could start by using network=rcn on those routes that run along the network of LCN+ links, provided that the cycle routes have definitely been developed, with the network=rcn added to the LCN route's relation itself rather than to LCN+ link(s) relations. To map this fully as a network, it might still need to include unnumbered LCN routes that follow LCN+ links. --MacLondon (talk) 01:20, 17 August 2018 (UTC)
Hi, at https://lists.openstreetmap.org/pipermail/talk-gb-london/2018-October/000175.html I have been outlining a rough idea of how we could deal with LCN since are quite a lot of overlaps. e.g. the CS1 and the LCN 10. --Ika-chan! (talk) 12:21, 12 October 2018 (UTC)
Well if you read the discussions above, you'll see we've already discussed all that here at some length... I don't think Quietways should be "local" routes - if they were tagged like that, there'd be nothing to distinguish them from the (literally hundreds) of local routes all over greater London, many of which only run for like a quarter of a mile and are waymarked by little else than a bicycle icon on the carriageay. Quietway routes usually traverse multiple boroughs, are well signed, and have a noticeably higher standard of infrastructure. - feline1
I apologise for falling behind on the discussion, but I was also thinking about retiring sections of LCN duplicated by CS and Q routes. I notice that the LCN signs go when the CS and Q routes arrive to take their place. I am open to accepting Q routes remaining at regional level. --Ika-chan! (talk) 18:41, 12 October 2018 (UTC)
No need to apologise, it's great to have more people interested! :) I'm afraid though, I don't think it's as simple as just retiring the LCN routes. To (try to!) summarise what we've been mulling around above - different boroughs take different attitudes to cycle signing. For instance, I live in Lewisham. Next to none of their LCN routes were ever signed with route numbers in the first place (just generic blue cycle signs). But next door in Greenwich, all the LCN routes are meticulously signed with route numbers. Some boroughs definitely actively take down old LCN route signs and carriageway markings when 'upgrading' to Quietways (eg I have seen this happen with Southwark) ... others never signed LCN routes properly in the first place so it's hard to know - Greenwich on the other hand, has just a couple of months ago marked the carriageway with a new Q route, but very deliberately refurbished all the old LCN17 route signs on it! And when they unveleiled Q14 on the Thames path, they included Sustrans National Cycle Route numbers on all the signs too.... but many other boroughs or TfL don't do that (for example, most of CS3 shares the route of Sustrans NCN13, but the CS3 signs ignore this). So, in some cases, LCN routes are being refurbished, not replaced... in others I feel their existence is basically gone (e.g. CS1 - there is no trace of it ever being LCN10). However LCN10 extends way way beyond the end of CS1 (all the way to the M25!) - so it still exists. And someone might still legitimately want to try cycling from, say, one end of a long LCN route to the other, even if there's a long bit in the middle that is now a CS or Q... It's complicated :) What I have done in some instances is made a new relation for bits of LCNs which really seem gone, and given them the tag "route=historic". That way they don't clutter up most rendered maps so much. But they could still easily be included in a custom map. A futher complication is that there appears to have been a lot more "London Cycle Network" than the numbered LCN routes - there's a lot of bits marked and signed on the carriageway with just generic "London Cycle Network" signs... and then as I was saying, there's plenty of numbered LCN routes which we know about from maps and planning documents, but which just have generic signage on the ground... but these are still different to borough's "preferred routes for cyclists" streets that might have a bit of traffic calming etc. --Feline1 (talk) 19:03, 12 October 2018 (UTC)