Re: [Talk-GB] Cycling, the law and traffic signs

2012-05-16 Thread Shaun McDonald

On 16 May 2012, at 01:05, Jason Cunningham wrote:

 On 15 May 2012 23:32, rob.j.nicker...@gmail.com wrote: 
 
 As I am not a regular cyclist I must admit that I don't pay much attention to 
 these signs. So my question is do Local Authorities use the cycle and foot 
 signs (segregated or otherwise) and reserve the cycle sign for cases where 
 traffic regulation prevents foot access (in which case foot=no would be 
 correct), or is use mixed?
 
 Cheers,
 Rob
 
 Unless it's been recently changed. the Cycle Only sign could never 
 prohibit 'pedestrian access' because use of the sign is defined by the 
 Department for Transports Traffic Signs Manual (chapter 3) [1]. 
 
 The DFT guidance confirms the signs can be used for routes where cycles can 
 travel and all other vehicular traffic is prohibited. Therefore this sign 
 must not be used to prohibit pedestrian access. The Manual also points out 
 usefulness of a convenient footway or footpath to lure pedestrians away from 
 this intended 'cycle only' way.

I find the cycle only sign is used in cases where there is also a separate 
pavement, thus the pedestrians can use that. They can in some cases be used 
where there is no pavement and it's not recommended for pedestrians to go that 
route. If cyclists are allowed and pedestrians are prohibited then a separate 
no pedestrians sign will be used.

Shaun

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Bulk railway station changes

2012-05-16 Thread Richard Mann
IMO it's better to add something clear than to shoehorn something into a
generic tag. Especially if you end up with compound values. OK so they
could be parsed, but it's just making work (both processing and
maintaining). Better to have something unambiguous like national_rail=yes
and london_underground=yes.

On Wed, May 16, 2012 at 4:35 AM, AJ Ashton aj.ash...@gmail.com wrote:

 Hi Richard  everyone,

 This started off simply as an effort to improve our display London
 Underground stations using existing OSM data, but was scope-creeped
 into much more and apparently we messed up.

 We've found that the lack of familiar London Underground and National
 Rail icons is a particularly strong sticking point with people who
 would otherwise happily switch to OSM, which is partly why we chose to
 focus on it. The tagging for stations is not so consistent, and my
 blog post goes into details about how we attempt to account for this
 as much as possible at the import  rendering stages. However certain
 inconsistencies seemed simple enough to just fix in OSM.

 We saw network=National Rail tags already in use at various stations
 and didn't think continuing to use them would be an issue. The
 imports/mechanical edits policies didn't come to my mind because we
 started with just a handful of edits. Even though this obviously ended
 up turning into many more, I thought that things were being done quite
 manually and carefully. There were no scripts or bots used, but the
 error the Craig points out looks like the result of a very bulk and
 incorrect copy/paste (or something) so clearly there were problems
 here.

  ... something that might seem simple
  from afar actually turns out to be a bit more nuanced, but by giving
 careful
  consideration to the nuances, we're making what is hands-down the best
 map
  of the world. I hope we can have a similarly useful conversation about
 the
  stations too.

 I guess our excitement to make awesome maps tripped us up here.
 Richard pointed out specifically that 'the network=National Rail tag
 is of debatable value and relevance'. I'm curious about the details of
 why.

 We just went with what seemed to be an established tagging system (but
 I guess is actually not). I am interested to hear tagging ideas that
 would be both correct and useful for rendering a map with appropriate
 icon styles.

 AJ @ MapBox

 ___
 Talk-GB mailing list
 Talk-GB@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-gb

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Cycling, the law and traffic signs

2012-05-16 Thread Richard Mann
I doubt there are any instances in the UK where there's a TRO supporting a
No Pedestrians sign on a cycle track (welcome to be proved wrong!). The
possibility exists in the legislation, but you'd have to explicitly sign it
(the white-bike-on-blue-circle does not of itself exclude pedestrians in
the UK).

The confusion arises because the european standard is that that sign really
does exclude pedestrians (and is very commonly used in Germany, the
Netherlands, Belgium, Austria...), but well, we do things differently.

Richard

On Wed, May 16, 2012 at 8:52 AM, Shaun McDonald
sh...@shaunmcdonald.me.ukwrote:


 On 16 May 2012, at 01:05, Jason Cunningham wrote:

 On 15 May 2012 23:32, rob.j.nicker...@gmail.com wrote:


 As I am not a regular cyclist I must admit that I don't pay much
 attention to these signs. So my question is do Local Authorities use the
 cycle and foot signs (segregated or otherwise) and reserve the cycle sign
 for cases where traffic regulation prevents foot access (in which case
 foot=no would be correct), or is use mixed?

 Cheers,
 Rob


 Unless it's been recently changed. the Cycle Only sign could never
 prohibit 'pedestrian access' because use of the sign is defined by the
 Department for Transports Traffic Signs Manual (chapter 3) [1].

 The DFT guidance confirms the signs can be used for routes where cycles
 can travel and all other vehicular traffic is prohibited. Therefore this
 sign must not be used to prohibit pedestrian access. The Manual also points
 out usefulness of a convenient footway or footpath to lure pedestrians away
 from this intended 'cycle only' way.


 I find the cycle only sign is used in cases where there is also a separate
 pavement, thus the pedestrians can use that. They can in some cases be used
 where there is no pavement and it's not recommended for pedestrians to go
 that route. If cyclists are allowed and pedestrians are prohibited then a
 separate no pedestrians sign will be used.

 Shaun


 ___
 Talk-GB mailing list
 Talk-GB@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-gb


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Bulk railway station changes

2012-05-16 Thread Richard Fairhurst
AJ Ashton wrote:
 We've found that the lack of familiar London Underground and 
 National Rail icons is a particularly strong sticking point with 
 people who would otherwise happily switch to OSM, which is 
 partly why we chose to focus on it.

Absolutely. It does look really good. :)

 I guess our excitement to make awesome maps tripped us up 
 here. Richard pointed out specifically that 'the network=National 
 Rail tag is of debatable value and relevance'. I'm curious about 
 the details of why.

Sadly anything to do with our godforsaken privatised railway system is
always more complicated than it needs to be!

For me I think the most problematic aspect is that there are actually
several things that could be called networks, particularly in urban areas
which have PTEs (Passenger Transport Executives) or similar. For example,
there's Network West Midlands around Birmingham, Metro in West/South
Yorkshire, Merseytravel/Merseyrail in Liverpool, and so on. (London has its
own peculiarities.) So you end up with network=National Rail;Metro which
is nasty, and breaks most toolchains which don't understand multiple values
for one key.

It may lend itself to an ncn/rcn/lcn or nwn/rwn/lwn solution, or Richard M's
idea of using a distinct tag, or tagging station operators (e.g.
operator=First Great Western) and rendering based on a set of those. I'm
tempted to suggest a generic tag for any country's national railway system
(mainline=yes|no or somesuch), and then you could render based on this tag
and the UK polygon. Or indeed we could just go with network=National Rail
as a good enough solution. I'd be interested to hear what others think.

cheers
Richard



--
View this message in context: 
http://gis.19327.n5.nabble.com/Bulk-railway-station-changes-tp5708989p5709044.html
Sent from the Great Britain mailing list archive at Nabble.com.

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Cycling, the law and traffic signs

2012-05-16 Thread Andy Allan
On 16 May 2012 01:05, Jason Cunningham jamicu...@googlemail.com wrote:

 Unless it's been recently changed. the Cycle Only sign could never
 prohibit 'pedestrian access' because use of the sign is defined by the
 Department for Transports Traffic Signs Manual (chapter 3) [1].

 The DFT guidance confirms the signs can be used for routes where cycles can
 travel and all other vehicular traffic is prohibited. Therefore this sign
 must not be used to prohibit pedestrian access. The Manual also points out
 usefulness of a convenient footway or footpath to lure pedestrians away from
 this intended 'cycle only' way.

Interesting stuff. So from my research this morning, sign 955 (cycle
only) is used in two scenarios - on-carriageway, for things like false
one-way streets, and on off-carriageway routes.

The text of the guidance is:


CYCLE TRACKS AND ROUTES SHARED WITH
PEDESTRIANS

17.32 An off-road cycle track is indicated by the
sign to diagram 955, which means that the route
is for cycles only and all other vehicular traffic
is prohibited. As the route is not intended for
pedestrians, there should be a convenient footway
or footpath nearby. The sign should be provided
at the start of the cycle track and where the track
crosses roads used by other traffic. The signs may
also be used as repeaters along the route. [...]

17.33 Where a footway (forming part of a road) or
footpath (e.g. through a park) has been converted to
a route shared by pedestrians and cyclists, signs to
either diagram 956 or 957 are used. These prohibit
the use of the route by any other vehicles. The sign
to diagram 956 indicates an unsegregated route.
It should be located where the shared route begins
and must be used as a repeater, at regular intervals
(direction 11), to remind both pedestrians and
cyclists that pedal cycles can be legally ridden on
the footway or footpath. [...]



So while it's correct that 955 doesn't prohibit pedestrians, there's
still a clear difference in intent between 955 and 956 (unsegregated
shared ped/cycle). How do we capture the difference? After all, from a
pedestrian's point of view, you'll be a bit miffed if OpenStreetMap
treats 955 and 956 as identical and you keep getting routed down paths
not intended for pedestrians.

Cheers,
Andy

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


[Talk-GB] routing on the road network

2012-05-16 Thread Tim Pigden
Hi
This is a general question about the current state of the maps.
When we use our commercial road networks (OS Mastermap  Navteq) for road
routing we tend to assume that the roads have been analysed for
connectivity, that there are no one-way streets leading to dead ends, that
you can't turn left off a flyover onto the road underneath and so forth).
we also assume that there is only one link connecting two points if there
is only one physical road link, and that all intersections are proper
intersections.

We have written tools to address check these issues in the past (and
clearly not all map makers have always addressed them internally).

How suitable is OSM GB for routing, right now, with this level of detail?
Do corresponding network analysis tools exist? I did try about 3 months ago
to follow details for building a routable network using pgrouting and one
of the programs in the chain  seemed to do thousands of fixups to the data.

If we did such an analysis (which would be quite a big investment) and came
to the conclusion there were 10s of thousands of errors - which seems to be
entirely possible - would there be any appetite in the community for fixing
them? Certainly such an endeavour would be way beyond our budget.

Tim

-- 
Tim Pigden
Optrak Distribution Software Limited
+44 (0)1992 517100
http://www.linkedin.com/in/timpigden
http://optrak.com
Optrak Distribution Software Ltd is a limited company registered in England
and Wales.
Company Registration No. 2327613 Registered Offices: Orland House, Mead
Lane, Hertford, SG13 7AT England
This email and any attachments to it may be confidential and are intended
solely for the use of the individual to whom it is addressed. Any views or
opinions expressed are solely those of the author and do not necessarily
represent those of Optrak Distribution Software Ltd. If you are not the
intended recipient of this email, you must neither take any action based
upon its contents, nor copy or show it to anyone. Please contact the sender
if you believe you have received this email in error.
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] routing on the road network

2012-05-16 Thread Rob Nickerson
Hi Tim,

I expect that you would find a good appetitie to fix any problems if we can
see that this will be highly valued (i.e. used in a great tool), and the
bug reports are simple to understand. I'll leave comments about the quality
to others, suffice to say that in the Midlands the roads and junctions are
mapped well, however turn restrictions are the missing component.

Regards,
Rob
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] routing on the road network

2012-05-16 Thread Frederik Ramm

Hi,

On 05/16/2012 11:56 AM, Tim Pigden wrote:

that there are no one-way streets leading to dead ends,


This is not common in OSM but I am not aware of anyone doing a network 
analysis that would fix such a problem.



that you can't turn left off a flyover onto the road underneath


In OSM, turning is only possible at nodes, not where two roads happen do 
cross.


Some editors and existing QA tools will whine when you have a non-noded 
intersection between two roads and neither of them is marked a bridge or 
tunnel.



we also assume that there is only one link connecting two points
if there is only one physical road link,


Opinions as to what makes up a physical road link may differ, especially 
when pavements or cycle lanes are concerned. In some countries, people 
map pavements and cycle lanes as separate geometries (eg 
http://www.openstreetmap.org/?lat=64.138954lon=-21.922888zoom=18layers=M).



and that all intersections are
proper intersections.


In the GIS world, a proper intersection is often one that is not 
traversed by any geometry, i.e. you will have roads beginning/ending 
there but not going through it. In OSM this is different; having a way 
traverse an intersection is normal and not considered an error.



How suitable is OSM GB for routing, right now, with this level of
detail? Do corresponding network analysis tools exist?


Tools exist but not in one central location. Some potential problems are 
caught by editors. Then there's OSM Inspector (tools.geofabrik.de/osmi) 
which has the Skobbler-sponsored routing layer that finds un-connected 
roads; keepright.at has something similar and also checks for suspect 
things like service roads branching off of motorways. Powerful 
project-osrm.org routing engine has the potential to be used as an 
analysis tool but doesn't produce any automated reports. The restriction 
analyzer at http://osm.virtuelle-loipe.de/restrictions/ specializes in 
debugging turn restrictions, and I'm sure there will be more tools I 
haven't mentioned!



I did try about 3
months ago to follow details for building a routable network using
pgrouting and one of the programs in the chain  seemed to do thousands
of fixups to the data.


Most likely the program broke up the roads into segments starting/ending 
at intersections, something that might be a required preprocessing step 
but nobody would suggest doing that in OSM. OSM is pretty suitable for 
routing in my opinion.



If we did such an analysis (which would be quite a big investment) and
came to the conclusion there were 10s of thousands of errors - which
seems to be entirely possible - would there be any appetite in the
community for fixing them? Certainly such an endeavour would be way
beyond our budget.


Appetite, yes, but you can also easily chase people away if your system 
detects too many things where people don't think it's a bug at all, so 
some tuning might be necessary. One of the weaknesses of most of the 
existing systems (with the exception of checks in the editor) is that 
once you fixed the bug, it might take hours or even days for the bug to 
be properly closed. A system that allows users to hit some kind of 
re-check this, I've just fixed it button would surely create more 
enthusiasm for cleaning up your area than most of what we currently have.


Bye
Frederik

--
Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09 E008°23'33

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Bulk railway station changes

2012-05-16 Thread SomeoneElse

Richard Fairhurst wrote:
Or indeed we could just go with network=National Rail as a good 
enough solution. 


My issue with National Rail was that, to me, (as I explained to the 
Peruvian chap who's edited Mansfield Woodhouse station):


National Rail means these people: http://www.nationalrail.co.uk/ , 
also http://www.atoc.org/


It's just an industry association of the various Train Operating 
Companies. They don't own, operate or have any direct involvement with 
the British rail _network_.



The only place I've heard national rail* used is in London to refer to 
non-underground stations (and even there, you still here British Rail 
station).   Everyone else says Railway.



I don't think that network=network_rail works either, as there will 
inevitably by issues in London where NR works on infrastructure for TfL 
Overground services.


So network=railway for me, since that's probably the best description 
of what it actually is.


Cheers,
Andy

* in lower case, where national simply means non-underground, and is 
a description rather than a name.



___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] routing on the road network

2012-05-16 Thread Philip Barnes
On Wed, 2012-05-16 at 13:42 +0200, Frederik Ramm wrote:

 Appetite, yes, but you can also easily chase people away if your system 
 detects too many things where people don't think it's a bug at all, so 
 some tuning might be necessary. One of the weaknesses of most of the 
 existing systems (with the exception of checks in the editor) is that 
 once you fixed the bug, it might take hours or even days for the bug to 
 be properly closed. A system that allows users to hit some kind of 
 re-check this, I've just fixed it button would surely create more 
 enthusiasm for cleaning up your area than most of what we currently have.
 
+1
I do go through the mapdust bugs reported by scrobbler users, many are
routing errors. One of the big problems is that I can see my map changes
almost immediatly but have to wait days before the routers start to use
the changes and I can test my changes and hence close the bug.

Phil


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Bulk railway station changes

2012-05-16 Thread SomeoneElse

AJ Ashton wrote:

.. and apparently we messed up.


...
AJ @ MapBox 


Sorry, but who's we here?  Is it a bunch of people at some other 
mailing list/forum, or who work for Mapbox, or something else?


Last night I spotted changes from someone (I think*) from Peru, and (I 
think) from the US - were these co-ordinated, did (a) just follow (b) 
blindly**, or what?


For the avoidance of doubt, I'm sure that everyone's trying to do the 
right thing here.  Just a question of what and how.


Cheers,
Andy

* now that everyone now uses gmail addresses it's difficult to tell 
which side of the planet anyone's on.


** I've seen naptan-coded bus stops in Germany


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Rights of way - Image vote

2012-05-16 Thread Andrew Chadwick
On 12/05/12 13:02, Philip Barnes wrote:

 They do vary between highway authorities, but well worth getting some
 photos of samples. The one thing waymarks have in common, and I can only
 claim knowledge of England and Wales here is that a public footpath has
 yellow arrows, public bridleways have blue arrows and the hardest to
 find of all are red arrows, used on B.O.T.A.Ts.

Not a waymarker, but the signposts are fairly rare too; Public Byway
or just Byway is the normal wording:

  https://wiki.openstreetmap.org/wiki/File:UK_Public_Byway_signpost.jpg
  http://osm.org/go/eutqlptvf--?m

and I don't think we could expect the waymarkers to say any more.


Predictably enough, the thin little road the one above points at is
blocked off at one end for larger vehicles:

  https://imgur.com/Tx9hI

To complicate matters further, that's a No Motor Vehicles sign under the
graffiti which presumably reflects a TRO filed somewhere in the bowels
of the local town hall. It's only applicable to the plugged end. A sign
on the far end warns of there being no sane turning places.

So it's not open to all traffic at all, and the sign doesn't call it
open to all traffic, but it should be tagged
designation=byway_open_to_all_traffic anyway :D

-- 
Andrew Chadwick

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Bulk railway station changes

2012-05-16 Thread Richard Mann
National Rail is what ATOC came up with to describe things that are
represented by the double-arrow symbol, and which would formerly have been
referred to as British Rail or informally as the rail network. (The staff
refer to it as the railway, but that's another subject)

National Rail isn't a great name, but it's the correct one. The symbol is
owned by ATOC.

On Wed, May 16, 2012 at 1:00 PM, SomeoneElse li...@mail.atownsend.org.ukwrote:

 Richard Fairhurst wrote:

 Or indeed we could just go with network=National Rail as a good enough
 solution.


 My issue with National Rail was that, to me, (as I explained to the
 Peruvian chap who's edited Mansfield Woodhouse station):

 National Rail means these people: http://www.nationalrail.co.uk/ , also
 http://www.atoc.org/

 It's just an industry association of the various Train Operating
 Companies. They don't own, operate or have any direct involvement with the
 British rail _network_.


 The only place I've heard national rail* used is in London to refer to
 non-underground stations (and even there, you still here British Rail
 station).   Everyone else says Railway.


 I don't think that network=network_rail works either, as there will
 inevitably by issues in London where NR works on infrastructure for TfL
 Overground services.

 So network=railway for me, since that's probably the best description of
 what it actually is.

 Cheers,
 Andy

 * in lower case, where national simply means non-underground, and is a
 description rather than a name.



 __**_
 Talk-GB mailing list
 Talk-GB@openstreetmap.org
 http://lists.openstreetmap.**org/listinfo/talk-gbhttp://lists.openstreetmap.org/listinfo/talk-gb

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Rights of way - Image vote

2012-05-16 Thread SomeoneElse

Andrew Chadwick wrote:
Not a waymarker, but the signposts are fairly rare too; Public Byway 
or just Byway is the normal wording


FWIW Derbyshire seem to have started (within the last 3 months or so) 
using Byway Open to All Traffic in full on signs.


Cheers,
Andy


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] routing on the road network

2012-05-16 Thread rob . j . nickerson
As noted there are some tools available to help find potential errors that  
effect routing. Have a look on the QA page [1], specifically at keepright,  
OSM Inspector and MapDust. As for editing OSM and proposing new tags,  
OpenStreetMap works a bit like wikipedia - we have a very flat structure  
and work together as a community to tackle issues such as whether new tags  
are needed or not. Discussions can get a little bit lengthy, but I believe  
we end up with a better solution in the end.


Regards,
Rob

[1] http://wiki.openstreetmap.org/wiki/Quality_Assurance



On , Tim Pigden tim.pig...@optrak.com wrote:
Error reporting would definitely be a challenge.Are there existing  
facilities to add suspect type tags to enable OSM itself to be the  
primary reporting medium? I haven't looked into the details of editing  
OSM data but adding new tags seems to require a collective decision.




(BTW don't get too excited anyone, this is definitely a long-term  
project).




Tim




On 16 May 2012 11:41, Rob Nickerson rob.j.nicker...@gmail.com wrote:




Hi Tim,


I expect that you would find a good appetitie to fix any problems if we  
can see that this will be highly valued (ie used in a great tool), and  
the bug reports are simple to understand. I'll leave comments about the  
quality to others, suffice to say that in the Midlands the roads and  
junctions are mapped well, however turn restrictions are the missing  
component.





Regards,
Rob




___



Talk-GB mailing list



Talk-GB@openstreetmap.org



http://lists.openstreetmap.org/listinfo/talk-gb








--
Tim Pigden
Optrak Distribution Software Limited
+44 (0)1992 517100
http://www.linkedin.com/in/timpigden



http://optrak.com
Optrak Distribution Software Ltd is a limited company registered in  
England and Wales.
Company Registration No. 2327613 Registered Offices: Orland House, Mead  
Lane, Hertford, SG13 7AT England


This email and any attachments to it may be confidential and are intended  
solely for the use of the individual to whom it is addressed. Any views  
or opinions expressed are solely those of the author and do not  
necessarily represent those of Optrak Distribution Software Ltd. If you  
are not the intended recipient of this email, you must neither take any  
action based upon its contents, nor copy or show it to anyone. Please  
contact the sender if you believe you have received this email in error.






___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Bulk railway station changes

2012-05-16 Thread Chris Hill

On 16/05/12 04:35, AJ Ashton wrote:

Hi Richard  everyone,

This started off simply as an effort to improve our display London
Underground stations using existing OSM data, but was scope-creeped
into much more and apparently we messed up.

We've found that the lack of familiar London Underground and National
Rail icons is a particularly strong sticking point with people who
would otherwise happily switch to OSM, which is partly why we chose to
focus on it. The tagging for stations is not so consistent, and my
blog post goes into details about how we attempt to account for this
as much as possible at the import  rendering stages. However certain
inconsistencies seemed simple enough to just fix in OSM.

We saw network=National Rail tags already in use at various stations
and didn't think continuing to use them would be an issue. The
imports/mechanical edits policies didn't come to my mind because we
started with just a handful of edits. Even though this obviously ended
up turning into many more, I thought that things were being done quite
manually and carefully. There were no scripts or bots used, but the
error the Craig points out looks like the result of a very bulk and
incorrect copy/paste (or something) so clearly there were problems
here.


... something that might seem simple
from afar actually turns out to be a bit more nuanced, but by giving careful
consideration to the nuances, we're making what is hands-down the best map
of the world. I hope we can have a similarly useful conversation about the
stations too.

I guess our excitement to make awesome maps tripped us up here.
Richard pointed out specifically that 'the network=National Rail tag
is of debatable value and relevance'. I'm curious about the details of
why.

We just went with what seemed to be an established tagging system (but
I guess is actually not). I am interested to hear tagging ideas that
would be both correct and useful for rendering a map with appropriate
icon styles.


I think talking to local mappers before making remote changes over a 
whole country is not just a good idea, but should be mandatory and 
failure to do so should routinely be met with a revert.


I accept that your motives were good.

The railways in the UK are a mix of nationalised and private companies 
who operate train services on tracks they don't own, using stations they 
might operate or might not, who sometimes get represented publicly by a 
company who doesn't run trains, operate stations or own any track. Some 
bigger stations are operated by the national track operator but most are 
operated by train operators, even though other train operators will 
often share the station too. In some cites there are added complications 
of underground, light rail and trams. In addition there are privately 
owned and run heritage lines, some of which provide local commuter 
services as well as tourist services.


You could use the well-established operator tag to help to distinguish 
the stations. It would take research to find them all and there will be 
a longish list to wade through for rendering. Maybe some UK mappers 
could help you with their local stations.


--
Cheers, Chris
user: chillly


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] routing on the road network

2012-05-16 Thread Andy Allan
On 16 May 2012 12:42, Frederik Ramm frede...@remote.org wrote:
 On 05/16/2012 11:56 AM, Tim Pigden wrote:

 that there are no one-way streets leading to dead ends,


 This is not common in OSM but I am not aware of anyone doing a network
 analysis that would fix such a problem.

Keepright has this check - dead-ended one-ways

Cheers,
Andy

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] routing on the road network

2012-05-16 Thread Andy Allan
I don't think the email below made it to the list:

 On , Tim Pigden tim.pig...@optrak.com wrote:
 Error reporting would definitely be a challenge.Are there existing
 facilities to add suspect type tags to enable OSM itself to be the primary
 reporting medium? I haven't looked into the details of editing OSM data but
 adding new tags seems to require a collective decision.

Please don't add bug reports to the OSM database itself, whether
through suspect tags or similar. I expect anything like that would
lead to getting blocked pretty quickly!

There's a variety of existing bug reporting / QA toolchains -
Keepright, OSB, Mapdust - for everything from auto-generated
calculations to end-user reports. It's best to pick one of them, and
add your additional insights to that.

Cheers,
Andy

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Bulk railway station changes

2012-05-16 Thread AJ Ashton
Thanks for the explanations of the complicated 'network' situation.

Richard Fairhurst wrote:
 It may lend itself to an ncn/rcn/lcn or nwn/rwn/lwn solution, ...

 I'm tempted to suggest a generic tag for any country's national railway system
 (mainline=yes|no or somesuch), and then you could render based on this tag
 and the UK polygon. ...

Either of these seem like good generic approaches. I'm sure there are
numerous complications and caveats to discuss, but a relatively
simple, global system is likely to have advantages for many
applications of the data.

 ... or tagging station operators (e.g.
 operator=First Great Western) and rendering based on a set of those.

Based on everyone's comments I'm leaning toward this as an alternative
approach to start with. It sounds like data/tags are less ambiguous
and already well-used (though not without oddities to be careful of).
On the icon rendering side it's slightly more complicated, but totally
manageable within our current rendering setup. The token-replacement
feature for image paths in in Mapnik 2 is really fantastic.

SomeoneElse wrote:
 Sorry, but who's we here? ...

We is MapBox, or at least the few MapBox employees who made these
edits. We're primarily based in the US, but also have some folks in
South America and Europe.

-- 
AJ Ashton

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


[Talk-GB] Cycle lanes and Cycle Tracks - how to map

2012-05-16 Thread rob . j . nickerson

Hi All,

I would like to improve the guidance given on the UK Tagging Guidelines  
page in regards to how to map cycle paths. As a non-cyclist I would like  
some advise. So far my research has found:


1. Definitions:
* Cycle Lane - lane marked out by painted lines _within_ the carriageway.  
May be mandatory or advisory.
* Cycle Track - a route other than within a carriageway - eg on a footway  
(legal term for pavement, rather than OSM highway=footway) adjacent to a  
carriageway, adjacent to the carriageway but separate from the footway  
(pavement), or on a route completely separate from a highway (eg a path  
through a park).


2. OSM tags:
In OSM we have highway=cycleway and we can also add cycleway=lane /  
cycleway=track / etc to any ways marked as highway=*.


3. Mapping practices:
Clearly a cycle lane should be tagged by adding cycleway=lane to the way  
represented by highway=*. Furthermore any cycle tracks that are on a  
route completely separate from a highway can be tagged as highway=cycleway  
(or highway=path, but lets shelve the Classic vs Alternative discussion for  
the moment).


This leaves cycle tracks that run alongside a highway but are not within  
the carriageway. How should they be tagged? Options are:


i) As a separate highway=cycleway (or path) with links back to the  
neighbouring roads whenever there is a 'connection' (eg a dropped kerb).

ii) Using cycleway=track on the highway=*.
iii) Both.

Advantages of (i) over (ii):
* When the cycle track is not within the carriageway you essentially have  
to decide whether to use it or not. Unlike lanes on a road you cannot  
simply switch back and forth easily due to kerbs.
* Can better represent the route of the cycle track (eg navigating over  
side roads).
* Will render easily without having to add complex rules to your rendering  
system.

* Perhaps easier to explain to a newcomer.
* Harder to be accidentally extended further than the cycle track actually  
goes.


Advantages of (ii) over (i):
* We do not tend to map individual pavements so why do it when there is a  
cycle track on them

* May appear less 'cluttered'
* Less work as there is no need to draw a separate way.
* Fewer ways and connecting nodes may make it easier to maintain and less  
prone to damage.
* Easier for routing software (for option i some rules will need to be  
added to tell the routing engine to prefer cycle tracks over adjacent roads  
/ a bicycle access tag would have to be added to the road to deter use).



So, over to you - thoughts? Statistics about current use in the UK?

Cheers,
Rob
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


[Talk-GB] Byways (Was: Rights of way - Image vote)

2012-05-16 Thread Rob Nickerson
I'm pretty happy to add that any way signposted as either a Byway Public
Byway or Byway Open to All Traffic should be tagged as
designation=byway_open_to_all_traffic as long as it also has a red arrow.
Unlike footpaths there is little chance of a landowner putting up a sign
for a byway unless it is a public right of way. There can also be no
confusion caused by long distance routes (as with footpaths).

Cheers,
Rob

On 12/05/12 13:02, Philip Barnes wrote:

* They do vary between highway authorities, but well worth getting some** 
photos of samples. The one thing waymarks have in common, and I can only** 
claim knowledge of England and Wales here is that a public footpath has** 
yellow arrows, public bridleways have blue arrows and the hardest to** find 
of all are red arrows, used on B.O.T.A.Ts.*
Not a waymarker, but the signposts are fairly rare too; Public Byway
or just Byway is the normal wording:

  https://wiki.openstreetmap.org/wiki/File:UK_Public_Byway_signpost.jpg
  http://osm.org/go/eutqlptvf--?m

and I don't think we could expect the waymarkers to say any more.


Predictably enough, the thin little road the one above points at is
blocked off at one end for larger vehicles:

  https://imgur.com/Tx9hI

To complicate matters further, that's a No Motor Vehicles sign under the
graffiti which presumably reflects a TRO filed somewhere in the bowels
of the local town hall. It's only applicable to the plugged end. A sign
on the far end warns of there being no sane turning places.

So it's not open to all traffic at all, and the sign doesn't call it
open to all traffic, but it should be tagged
designation=byway_open_to_all_traffic anyway :D

--
Andrew Chadwick
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


[Talk-GB] Cycling, the law and traffic signs

2012-05-16 Thread Ralph Smyth
I think the confusion here relates to a failure to differentiate
criminal law - in this case failing to comply with a traffic sign - and
civil law - trespass - in this case not being within the class(es) of
users permitted on a particular section of highway.
 
It's not a criminal offence to cycle on a footpath (as opposed to a
footway along the side of a road) but there can still be a bicycle=no
tag. Similarly with a cycle (only) track there could be a foot=no even
if someone walking there would not be committing a crime as they could
still be technically trespassing, which might be relevant if there was a
collision between them and someone cycling.
 
Off the top of my head the no pedestrian sign is simply signifying
another restriction (such as the prohibiting being a pedestrian on a
motorway unless walking from a broken down vehicle) since there's no
equivalent provision to section 36 of the Road Traffic Act requiring
compliance with traffic signs for anyone other than those driving or
propelling a vehicle. 
 
 
Ralph
 
winmail.dat___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Cycle lanes and Cycle Tracks - how to map

2012-05-16 Thread Andrew Chadwick
On 16/05/12 16:55, rob.j.nicker...@gmail.com wrote:

 I would like to improve the guidance given on the UK Tagging Guidelines
 page in regards to how to map cycle paths. As a non-cyclist I would like
 some advise. So far my research has found:
 
 1. Definitions:
 * Cycle Lane - lane marked out by painted lines _within_ the
 carriageway. May be mandatory or advisory.
 * Cycle Track - a route other than within a carriageway - e.g. on a
 footway (legal term for pavement, rather than OSM highway=footway)
 adjacent to a carriageway, adjacent to the carriageway but separate from
 the footway (pavement), or on a route completely separate from a highway
 (e.g. a path through a park).

Looks like a good start. Cycle Track is also the legal term. Would
designation coding be appropriate, assuming designation_codes are
derivable from official sources? On pavements, there's a distinction
cyclists and sign-makers make between segregated (from pedestrians,
paint or low kerb divider) and shared-use (shared with pedestrians, no
dividers). Any more?

 2. OSM tags:
 In OSM we have highway=cycleway and we can also add cycleway=lane /
 cycleway=track / etc to any ways marked as highway=*.

There's a wealth of discussion out there regarding how to tag
side-specific lanes or tracks. Are cycleway:left=* and cycleway:right=*
winning out?

 3. Mapping practices:
 Clearly a cycle lane should be tagged by adding cycleway=lane to the
 way represented by highway=*. Furthermore any cycle tracks that are on
 a route completely separate from a highway can be tagged as
 highway=cycleway (or highway=path, but lets shelve the Classic vs
 Alternative discussion for the moment).
 
 This leaves cycle tracks that run alongside a highway but are not
 within the carriageway. How should they be tagged? Options are:
 
 i) As a separate highway=cycleway (or path) with links back to the
 neighbouring roads whenever there is a 'connection' (e.g. a dropped kerb).
 ii) Using cycleway=track on the highway=*.
 iii) Both.

Personally, I'd say either (i) or (ii); both are valid, provide lots of
rendering options, and should route correctly provided both systems link
up. (i) is most appropriate in places with good aerial imagery, and
solves the side-of-road issue neatly. Also it should be done in areas where

  https://wiki.openstreetmap.org/wiki/Tag:footway%3Dsidewalk
  https://wiki.openstreetmap.org/wiki/Tag:footway%3Dcrossing

are being attempted, just for continuity. (ii) is a decent enough
approach even in micromapped areas as a first stab at getting a track
onto the map, particularly if you do the :left or :right thing.

However (iii) seems like mapping the same thing twice, which is
generally a bad idea:

  https://wiki.openstreetmap.org/wiki/One_feature,_one_OSM_element

-- 
Andrew Chadwick

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Byways (Was: Rights of way - Image vote)

2012-05-16 Thread Richard Mann
You sometimes get a simple direction sign at a road junction saying
Byway. It just means it doesn't go anywhere very much, but otherwise it's
a normal unclassified (non-urban) road.

Richard

On Wed, May 16, 2012 at 5:43 PM, Rob Nickerson rob.j.nicker...@gmail.comwrote:



 I'm pretty happy to add that any way signposted as either a Byway
 Public Byway or Byway Open to All Traffic should be tagged as
 designation=byway_open_to_all_traffic as long as it also has a red arrow.
 Unlike footpaths there is little chance of a landowner putting up a sign
 for a byway unless it is a public right of way. There can also be no
 confusion caused by long distance routes (as with footpaths).

 Cheers,
 Rob

 On 12/05/12 13:02, Philip Barnes wrote:
 
 * They do vary between highway authorities, but well worth getting some** 
 photos of samples. The one thing waymarks have in common, and I can 
 only** claim knowledge of England and Wales here is that a public 
 footpath has** yellow arrows, public bridleways have blue arrows and the 
 hardest to** find of all are red arrows, used on B.O.T.A.Ts.*
 Not a waymarker, but the signposts are fairly rare too; Public Byway
 or just Byway is the normal wording:
 
   https://wiki.openstreetmap.org/wiki/File:UK_Public_Byway_signpost.jpg
   http://osm.org/go/eutqlptvf--?m
 
 and I don't think we could expect the waymarkers to say any more.
 
 
 Predictably enough, the thin little road the one above points at is
 blocked off at one end for larger vehicles:
 
   https://imgur.com/Tx9hI
 
 To complicate matters further, that's a No Motor Vehicles sign under the
 graffiti which presumably reflects a TRO filed somewhere in the bowels
 of the local town hall. It's only applicable to the plugged end. A sign
 on the far end warns of there being no sane turning places.
 
 So it's not open to all traffic at all, and the sign doesn't call it
 open to all traffic, but it should be tagged
 designation=byway_open_to_all_traffic anyway :D
 
 --
 Andrew Chadwick



 ___
 Talk-GB mailing list
 Talk-GB@openstreetmap.org
 http://lists.openstreetmap.org/listinfo/talk-gb


___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Cycle lanes and Cycle Tracks - how to map

2012-05-16 Thread Steve Dobson


On 16/05/12 17:55, Andrew Chadwick wrote:
 3. Mapping practices:
 Clearly a cycle lane should be tagged by adding cycleway=lane to the
 way represented by highway=*. Furthermore any cycle tracks that are on
 a route completely separate from a highway can be tagged as
 highway=cycleway (or highway=path, but lets shelve the Classic vs
 Alternative discussion for the moment).

 This leaves cycle tracks that run alongside a highway but are not
 within the carriageway. How should they be tagged? Options are:

 i) As a separate highway=cycleway (or path) with links back to the
 neighbouring roads whenever there is a 'connection' (e.g. a dropped kerb).
 ii) Using cycleway=track on the highway=*.
 iii) Both.
 
 Personally, I'd say either (i) or (ii); both are valid, provide lots of
 rendering options, and should route correctly provided both systems link
 up. (i) is most appropriate in places with good aerial imagery, and
 solves the side-of-road issue neatly. Also it should be done in areas where
 
   https://wiki.openstreetmap.org/wiki/Tag:footway%3Dsidewalk
   https://wiki.openstreetmap.org/wiki/Tag:footway%3Dcrossing
 
 are being attempted, just for continuity. (ii) is a decent enough
 approach even in micromapped areas as a first stab at getting a track
 onto the map, particularly if you do the :left or :right thing.

As a cyclist with a Garman linking the cycle{way/path}s to the
neighboring roads is essential.  I had to spend some time a while back
linking type cycleways to the roads around my town where they crossed
roads in order to get the routing to work on my Garman.

Steve
-- 
Steve Dobbo Dobson

___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Cycle lanes and Cycle Tracks - how to map

2012-05-16 Thread Richard Mann
Gosh, you are a glutton for punishment.

cycleway=track is used extensively in some countries
highway=cycleway is use extensively in some countries
cycleway=track was only rendered on OCM relatively recently
cycleway:left|right=track|lane isn't rendered on OCM
the Danes had a big argument about which to use and settled on
cycleway=track, despite it not being rendered on OCM
cycleway=track gives you more control over the rendering
highway=cycleway is easier to route, though unpacking cycleway=track isn't
difficult
sub-tagging of cycleways is difficult (eg their membership of a route
relation) if you use cycleway=track

In essence it comes down to the problem that recombining two parallel ways
in order to render them neatly is next-to-impossible. Whereas putting the
tags on a single way loses some micro-geography.

I'd go for cycleway=track, but I'm not prepared to go round deleting
highway=cycleway, and thus having lots of stuff disappear in OCM. So until
OCM can render cycleway:left|right properly, we're probably stuck with both.

Richard
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb


Re: [Talk-GB] Cycle lanes and Cycle Tracks - how to map

2012-05-16 Thread rob . j . nickerson

Haha, I only mean well :-)

The issue came about from trying to improve the guidance provided on the UK  
tagging guidelines. Currently I have copied over the guidance that already  
existed (to the consultation page), however this was very limited and has  
already had cries that foot=no is not correct.


Now I can easily improve this section to draw a distinction about  
cycleway=lane when the cycle path is _within_ the carriageway (with some  
images). However there are 2 methods for cycle tracks alongside a road but  
not within the carriageway. Question to talk-gb is do we in the UK have a  
preference? If not then we need to discuss this on the wiki page and stress  
that just map it regardless of non-consensus.


Note: One thought on left / right is that mappers have to realise that a  
road has a direction (as it is drawn as an arrow). From my conversations  
with new mappers this (and the similar Forward/Backward) is not understood  
as we tend to think in terms of northbound / southbound / east / west.


Cheers,
Rob

ps For transparency I am currently thinking that the highway=*   
cycleway=track combination is a good start but we should be aiming to map  
the cycle track as separate from the highway using a highway=cycleway (or  
path) way.




On , Richard Mann richard.mann.westoxf...@gmail.com wrote:

Gosh, you are a glutton for punishment.



cycleway=track is used extensively in some countries
highway=cycleway is use extensively in some countries
cycleway=track was only rendered on OCM relatively recently



cycleway:left|right=track|lane isn't rendered on OCM
the Danes had a big argument about which to use and settled on  
cycleway=track, despite it not being rendered on OCM

cycleway=track gives you more control over the rendering


highway=cycleway is easier to route, though unpacking cycleway=track  
isn't difficult
sub-tagging of cycleways is difficult (eg their membership of a route  
relation) if you use cycleway=track



In essence it comes down to the problem that recombining two parallel  
ways in order to render them neatly is next-to-impossible. Whereas  
putting the tags on a single way loses some micro-geography.



I'd go for cycleway=track, but I'm not prepared to go round deleting  
highway=cycleway, and thus having lots of stuff disappear in OCM. So  
until OCM can render cycleway:left|right properly, we're probably stuck  
with both.




Richard




___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb