Re: [OSM-talk-be] wiki highway conventions

2012-12-28 Thread Kevin Grossard

The Ambachtenlaan is the main road to get to the industrial site located at the 
dead end. It was designed to get extended. This ring should connect the 
Oplintersesteenweg, the N29 and the N223 to the R27 ring. Last plans designed a 
parallel ringroad with the R27a because this road can't be used as a new 
secondary road anymore due to the occurence of direct driveways to the 
roadshops and industrial plants.
i think it can be secondary as well as tertiairy.
> Message: 2
> Date: Fri, 28 Dec 2012 12:00:13 +0100
> From: Ben Laenen 
> To: talk-be@openstreetmap.org
> Subject: Re: [OSM-talk-be] wiki highway conventions
> Message-ID: <201212281200.13876.benlae...@gmail.com>
> Content-Type: Text/Plain; charset="iso-8859-15"
> 
> On Friday 28 December 2012 10:06:49 Kevin Grossard wrote:
> > I can understand the current classification but i prefere to use the
> > classification of the government. Should i undo my changement and make the
> > ring primary again? In that case it makes sense for me to upgrade the N223
> > to a primary road allthough it isn't in rule with the wiki. In the other
> > case i suggest to downgrade the N29 to a secondary road.
> 
> If it were only up to me, I'd classify N29 as primary (I've seen some parts 
> of 
> that road and it's certainly "big enough" for primary). N223 no idea how that 
> road looks like, but I'd keep it secondary. And then to have a "logically 
> connected" map, the N3/N29 in the center of Tienen should all be primary as 
> well. Routers should pick the R27 based on tags like maxspeed and other data 
> like the number of crossings.
> 
> But it's not only me who can decide of course :-)
> 
> btw, what's the status of the Ambachtenlaan, which is that dead-end secondary 
> road on the east of Tienen?
> 
> Ben
> 
> 
> 
> --
> 
> ___
> Talk-be mailing list
> Talk-be@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-be
> 
> 
> End of Talk-be Digest, Vol 60, Issue 23
> ***   
>   
___
Talk-be mailing list
Talk-be@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-be


Re: [OSM-talk-be] wiki highway conventions

2012-12-28 Thread Ben Laenen
On Friday 28 December 2012 10:06:49 Kevin Grossard wrote:
> I can understand the current classification but i prefere to use the
> classification of the government. Should i undo my changement and make the
> ring primary again? In that case it makes sense for me to upgrade the N223
> to a primary road allthough it isn't in rule with the wiki. In the other
> case i suggest to downgrade the N29 to a secondary road.

If it were only up to me, I'd classify N29 as primary (I've seen some parts of 
that road and it's certainly "big enough" for primary). N223 no idea how that 
road looks like, but I'd keep it secondary. And then to have a "logically 
connected" map, the N3/N29 in the center of Tienen should all be primary as 
well. Routers should pick the R27 based on tags like maxspeed and other data 
like the number of crossings.

But it's not only me who can decide of course :-)

btw, what's the status of the Ambachtenlaan, which is that dead-end secondary 
road on the east of Tienen?

Ben

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


Re: [OSM-talk-be] wiki highway conventions

2012-12-28 Thread Glenn Plas

On 12/28/2012 01:10 AM, Ben Laenen wrote:

On Friday 28 December 2012 00:22:17 Glenn Plas wrote:

But the general wiki (which is probably USA based if I'm not mistaking
here) does say so:

http://wiki.openstreetmap.org/wiki/Tag:highway%3Dprimary

There is not much room for interpretation except 'large' definition.

It's not really any country based. The definition is as broad as possible, and
every country makes its own rules.


Maybe I should have been more specific,  it's just that the person that 
wrote that would have done it from their own point of view and 
location.   Chances where that he/they come from the USA.   A lot of 
assumption of course.




But I understand if you drive over R22 you would classify it primary
(although it has 4 lanes divided by a central barrier, not 2). That
would be a trunk to me as it connect to the E19 complex in Vilvoorde.

There are cycleways, so no trunk...


Right again.  Forgot about the cycleway requirement for a second. The 
definition of the word 'trunk' got me there.




As a subnote, the N211 which belongs to that complex goes primary until
the city center, that is wrong imho matched against the theory.

Yeah, N211 should be secondary east of the R22.

West of the R22: makes sense as primary. Strictly following the rules it
should be secondary of course, but this is one of those possible exceptions.
I'll change that short section soon if we ok on this, but I think it's 
the other way around the east section of R22 at the roundabout should 
remain primary and the west part changed to secondary.




I guess someone did something similar in Vilvoorde with the R22. Whether
that should be done, not really sure here.

That's why I bring it up as I'm not totally sure myself, but when I'm
not sure, I check the wiki ...

And of course the wiki isn't exactly the best reference for it :-) The
"definition" has been slowly adjusted on the mailing list for example, but we
can't expect everyone to know the entire history of course...
Since I'm only subscribed recently but active much longer, I'm one of 
them that missed it.





I guess I'll make a small summary with some examples (limiting to
trunk/primary/secondary):


* trunk: "expressways", prohibited for pedestrians and cyclists. May be signed
with F9 (motorroad), but not always. Not all roads with F9 are trunks (e.g.
tunnels or bridges if the road it is part of isn't classified as trunk)

* primary: Nx or Nxx roads

* secondary: Nxxx roads (or P roads)


Exceptions to primary/secondary above:

* ringways (usually R roads, but can be N roads): classification from highest
classified road that connects to it. Inside R road primary road becomes
secondary if the ringway deviates the traffic between the *primary* roads.
Secondary roads stay secondary inside ringways (they don't become tertiary)
(*).
e.g. R13 http://osm.org/go/0ErVXu0

This should end up on the wiki, it's good info.




* everywhere it makes sense... When a road without number connects two parts
that do for example, or any other weird situation you may encounter.
e.g. Corbiestraat as primary http://osm.org/go/0ErYW5cL
Plezantstraat as secondary (no road number) http://osm.org/go/0Ej6EwLs--



Roads with suffix in their number (e.g. N123a): use your brains. Almost
every case is unique here so it's impossible to writ down a good rule.
e.g. N1c http://osm.org/go/0EpMnUUZ--
N60b http://osm.org/go/0EiFrVaz--
N15a http://osm.org/go/0EpLMAqA-
N42c http://osm.org/go/0EiwSJL_-


If a N-road really shouldn't be secondary at all, it can be classified
unclassief/residential (not tertiary, if it can be tertiary it's sufficient to
be secondary)
e.g. N408 http://osm.org/go/0EpSSDeg



(*) note that the previous years a lot of roads in city centers have been
transferred from the region to the municipalities, and they lost their road
numbers, so they've become tertiary even though there are probably still road
numbers found on the street.



I guess these are the rules we have now. They sometimes allow for some
discussion in some cases of course, but they do seem to create a nice map.


This should end up on the wiki, it's good info.   I do like the map as 
well as it is now, but I try not let the renderer influence my actions 
(too much) although it plays a role.


Glenn

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


Re: [OSM-talk-be] wiki highway conventions

2012-12-28 Thread Kevin Grossard

>On 12/27/2012 23:22 PM, Glenn Plas wrote:
>On 12/27/2012 10:28 PM, Ben Laenen wrote:
>> I also wanted to point at Tienen, but I see it has been recently changed by
>> someone:http://osm.org/go/0EqQDB8
>> There's the N29 coming in from the north, and the N3/N29 should stay primary
>> because there's no ringway on the north. Now someone changed it to secondary
>> which is incorrect IMO.
>
>It look to me someone hopes that this information ends up in GPS devices 
>so they router prefers the primary road :) .  I don't know the place 
>there but it looks kind of strange to me that the N3/N29 isn't  
>primary.  But zooming out and taking a peek at Leuven which I know 
>better I see the ring there is marked primary.   So looking that those 
>without any knowledge on Tienen,  I would conclude that the Tienen 
>N3/ring part is a lot different from the Leuven one, so it would support 
>the secondary classification.
>
>Thanks for the examples.
>
>Glenn

I recently changed the ring of Tienen (N3) to a secondary road because the 
Sint-Truidensesteenweg
(also N3) between the ring and the roundabout with the R27 already was a 
secondary road.
The consequence of this changement is that there is no primary road connected 
anymore with the
N29 Diestsesteenweg which is classified as a primary road (the only one left on 
the ring).
For me it also doesn't make sense to keep the N29 primary if the N223 is 
secondary.
They both connect the relative small cities Aarschot, Diest and the motorway 
E314 to another
relative small city Tienen and the motorway E40.
 
Nowdays the ring of Tienen is quite simular with the one of Leuven (southern 
part). Only the speed limit
in Tienen is set on 70 km/h in stead of 50 km/h in Leuven. There are also plans 
to reduce the small ring
of Tienen (N3) from a road with 2 lanes on each side to 1 lane on each side 
(already degrated to a
secondary road type 3 in the structure plan of Vlaams-Brabant).
An extension of the R27 to the N29 and the N223 won't be realised soon so there 
is no alternative
for the N3 and no reason to degradate it. So even if the primary road changes 
to a secondary one,
the gps will still folow the same route.
 
I can understand the current classification but i prefere to use the 
classification of the government.
Should i undo my changement and make the ring primary again? In that case it 
makes sense for
me to upgrade the N223 to a primary road allthough it isn't in rule with the 
wiki.
In the other case i suggest to downgrade the N29 to a secondary road.
 
Kevin ___
Talk-be mailing list
Talk-be@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-be


Re: [OSM-talk-be] wiki highway conventions

2012-12-28 Thread Johan C
2012/12/28 Ben Laenen 

> And of course the wiki isn't exactly the best reference for it :-) The
> "definition" has been slowly adjusted on the mailing list for example, but
> we
> can't expect everyone to know the entire history of course...
>

Too bad that useful information is more or less kept hidden for others
inside country specific mailing lists. If the wiki is no longer a good
reference, it should be updated. Volunteers?
___
Talk-be mailing list
Talk-be@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-be


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Ben Laenen
On Friday 28 December 2012 00:22:17 Glenn Plas wrote:
> But the general wiki (which is probably USA based if I'm not mistaking
> here) does say so:
> 
> http://wiki.openstreetmap.org/wiki/Tag:highway%3Dprimary
> 
> There is not much room for interpretation except 'large' definition.

It's not really any country based. The definition is as broad as possible, and 
every country makes its own rules.


> But I understand if you drive over R22 you would classify it primary
> (although it has 4 lanes divided by a central barrier, not 2). That
> would be a trunk to me as it connect to the E19 complex in Vilvoorde.

There are cycleways, so no trunk...


> As a subnote, the N211 which belongs to that complex goes primary until
> the city center, that is wrong imho matched against the theory.

Yeah, N211 should be secondary east of the R22.

West of the R22: makes sense as primary. Strictly following the rules it 
should be secondary of course, but this is one of those possible exceptions.


> > I guess someone did something similar in Vilvoorde with the R22. Whether
> > that should be done, not really sure here.
> 
> That's why I bring it up as I'm not totally sure myself, but when I'm
> not sure, I check the wiki ...

And of course the wiki isn't exactly the best reference for it :-) The 
"definition" has been slowly adjusted on the mailing list for example, but we 
can't expect everyone to know the entire history of course...


I guess I'll make a small summary with some examples (limiting to 
trunk/primary/secondary):


* trunk: "expressways", prohibited for pedestrians and cyclists. May be signed 
with F9 (motorroad), but not always. Not all roads with F9 are trunks (e.g. 
tunnels or bridges if the road it is part of isn't classified as trunk)

* primary: Nx or Nxx roads

* secondary: Nxxx roads (or P roads)


Exceptions to primary/secondary above: 

* ringways (usually R roads, but can be N roads): classification from highest 
classified road that connects to it. Inside R road primary road becomes 
secondary if the ringway deviates the traffic between the *primary* roads. 
Secondary roads stay secondary inside ringways (they don't become tertiary) 
(*).
e.g. R13 http://osm.org/go/0ErVXu0



* everywhere it makes sense... When a road without number connects two parts 
that do for example, or any other weird situation you may encounter.
e.g. Corbiestraat as primary http://osm.org/go/0ErYW5cL
Plezantstraat as secondary (no road number) http://osm.org/go/0Ej6EwLs--



Roads with suffix in their number (e.g. N123a): use your brains. Almost 
every case is unique here so it's impossible to writ down a good rule.
e.g. N1c http://osm.org/go/0EpMnUUZ--
N60b http://osm.org/go/0EiFrVaz--
N15a http://osm.org/go/0EpLMAqA-
N42c http://osm.org/go/0EiwSJL_-


If a N-road really shouldn't be secondary at all, it can be classified 
unclassief/residential (not tertiary, if it can be tertiary it's sufficient to 
be secondary)
e.g. N408 http://osm.org/go/0EpSSDeg



(*) note that the previous years a lot of roads in city centers have been 
transferred from the region to the municipalities, and they lost their road 
numbers, so they've become tertiary even though there are probably still road 
numbers found on the street.



I guess these are the rules we have now. They sometimes allow for some 
discussion in some cases of course, but they do seem to create a nice map.

Ben

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


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Glenn Plas

On 12/27/2012 10:28 PM, Ben Laenen wrote:

On Thursday 27 December 2012 22:05:17 Glenn Plas wrote:

There is a simple rule I use, which is being used for certain in plenty
of countries like UK and described in the OSM wiki.   A primary road is
a road that connects larger cities, so a N road between 2 small town
would be secondary.  I think it matters that you look at the whole road
with the same reference.  That is the cities you look at.

UK doesn't use a rule like that, they're even much more strict than us using
road numbers...


Yeah, it seems you're right, I've must have assumed at the time I was 
looking at the UK sections.  they have conventions based on road types 
(ref.) See 
http://wiki.openstreetmap.org/wiki/WikiProject_United_Kingdom_A_and_B_Roads#How_to_tag


But the general wiki (which is probably USA based if I'm not mistaking 
here) does say so:


http://wiki.openstreetmap.org/wiki/Tag:highway%3Dprimary

There is not much room for interpretation except 'large' definition.


How much traffic passes over it, doesn't really matter.   For instance,
the N1 Mechelen-Brussel, someone tagged the part through Vilvoorde as
secondary.  Since Mechelen, Vilvoorde and Brussel counts as a
connection through big cities plus, well ... It's number 1 of the
N-class roads.

I haven't really checked the situation in Vilvoorde, but one of the "rules" is
that within a city center, primary becomes secondary if there's a ringway that
will connect to the other primary roads to the city.
Well, I know the place pretty well and there is no real ring road around 
Vilvoorde that reconnects to the N1.   There are ways to get around 
there using Woluwelaan(R22) but that's not a common alternative 
practically.Your description however fits what I've changed as the 
Woluwelaan was marked as primary and it could indeed be considered a 
quarter ring (tops),  what that one connects aren't really big cities 
(Vilvoorde -> Diegem -> Zaventem) . But it doesn't easily connect back 
to the N1.


But I understand if you drive over R22 you would classify it primary 
(although it has 4 lanes divided by a central barrier, not 2). That 
would be a trunk to me as it connect to the E19 complex in Vilvoorde.


As a subnote, the N211 which belongs to that complex goes primary until 
the city center, that is wrong imho matched against the theory.

For example:
http://osm.org/go/0Eg4Fgm
N8 is secondary in Zwevegem, since the N391 was made to create a ringway
around Zwevegem.

This one makes perfect sense to be primary to me.


http://osm.org/go/0EpYdO2L
N1 crosses centre of Antwerp, but becomes secondary within the Singel.
I totally agree on this one too.  It still fits the 'primary' class to 
me as it connects cities, so once you are there, there isn't anything 
big you can connect any more.  Very nice example.  Haven't even thought 
of checking the other side of the N1.



I guess someone did something similar in Vilvoorde with the R22. Whether that
should be done, not really sure here.


That's why I bring it up as I'm not totally sure myself, but when I'm 
not sure, I check the wiki ...



I also wanted to point at Tienen, but I see it has been recently changed by
someone:http://osm.org/go/0EqQDB8
There's the N29 coming in from the north, and the N3/N29 should stay primary
because there's no ringway on the north. Now someone changed it to secondary
which is incorrect IMO.


It look to me someone hopes that this information ends up in GPS devices 
so they router prefers the primary road :) .  I don't know the place 
there but it looks kind of strange to me that the N3/N29 isn't  
primary.  But zooming out and taking a peek at Leuven which I know 
better I see the ring there is marked primary.   So looking that those 
without any knowledge on Tienen,  I would conclude that the Tienen 
N3/ring part is a lot different from the Leuven one, so it would support 
the secondary classification.


Thanks for the examples.

Glenn

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


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Ben Laenen
On Thursday 27 December 2012 22:05:17 Glenn Plas wrote:
> There is a simple rule I use, which is being used for certain in plenty
> of countries like UK and described in the OSM wiki.   A primary road is
> a road that connects larger cities, so a N road between 2 small town
> would be secondary.  I think it matters that you look at the whole road
> with the same reference.  That is the cities you look at.

UK doesn't use a rule like that, they're even much more strict than us using 
road numbers...


> How much traffic passes over it, doesn't really matter.   For instance,
> the N1 Mechelen-Brussel, someone tagged the part through Vilvoorde as
> secondary.  Since Mechelen, Vilvoorde and Brussel counts as a
> connection through big cities plus, well ... It's number 1 of the
> N-class roads.

I haven't really checked the situation in Vilvoorde, but one of the "rules" is 
that within a city center, primary becomes secondary if there's a ringway that 
will connect to the other primary roads to the city.

For example:
http://osm.org/go/0Eg4Fgm
N8 is secondary in Zwevegem, since the N391 was made to create a ringway 
around Zwevegem.

http://osm.org/go/0EpYdO2L
N1 crosses centre of Antwerp, but becomes secondary within the Singel.

I guess someone did something similar in Vilvoorde with the R22. Whether that 
should be done, not really sure here.

I also wanted to point at Tienen, but I see it has been recently changed by 
someone: http://osm.org/go/0EqQDB8
There's the N29 coming in from the north, and the N3/N29 should stay primary 
because there's no ringway on the north. Now someone changed it to secondary 
which is incorrect IMO.

Ben

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


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Glenn Plas

On 12/27/2012 11:03 AM, Ben Laenen wrote:

On Thursday 27 December 2012 09:51:36 Kevin Grossard wrote:

The wiki about the highway conventions distinguishes primairy, secondary
and tertiairy roads using the N-numbers (although there are some question
marks). Using the current conventions means using the old classification
when the N-numbers were given. Various N-ways got reconstructed, the
traffic got redirected by other roads, some roads aren't suitable anymore
for the current traffic.

The spatial structure plan for Flanders (ruimtelijk structuurplan
Vlaanderen) has a list with the primairy roads [pages 368-377
http://www2.vlaanderen.be/ruimtelijk/docs/rsv2011/RSV2011.pdf]. I suggest
to adjust the wiki so there will be an uniform highway classification
where primairy roads are the roads selected in the structure plan. The
other N-roads can be describes as secondary roads. Tertiary roads can be
described as other important local roads (nl: steenwegen die geen
secundaire weg zijn). Minor or residential roads can be understand as
local roads (nl: lokale wegen). What do you think?

I don't really think we can just take that classification and apply it to OSM.
It may not look like it from the map in that file, but it would make Flanders
almost void of primary roads. Our government isn't very keen to give lots of
roads a primary status. And all the dead-end primary roads won't give a nice
map either (ringway of Lier without any primary road connecting to it, things
like that).

I've been aware about the official classification (read some previous
discussions in the mailing list archive), but I think we at least need to
include some secundary classified roads tagged as primary in OSM to make a
useful map. So far, we don't have a list of those. In the UK, the category of
official primary roads would be tagged as trunk in OSM.

That said, using road numbers to determine OSM classification is actually how
it's done in most countries. It's not perfect by far but we've always allowed
some deviations from the rule where it makes sense.

Ben

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



There is a simple rule I use, which is being used for certain in plenty 
of countries like UK and described in the OSM wiki.   A primary road is 
a road that connects larger cities, so a N road between 2 small town 
would be secondary.  I think it matters that you look at the whole road 
with the same reference.  That is the cities you look at.


How much traffic passes over it, doesn't really matter.   For instance, 
the N1 Mechelen-Brussel, someone tagged the part through Vilvoorde as 
secondary.  Since Mechelen, Vilvoorde and Brussel counts as a  
connection through big cities plus, well ... It's number 1 of the 
N-class roads.  It should make sense that this is a primary road.  Even 
after redesigning the part through Vilvoorde, it didn't make any sense 
at all to keep this secondary.  This road should be primary from start 
to finish (Antwerpen-Brussel).  I live closeby and Vilvoorde is being 
reconstructed at a fast pace, I already added a extra roundabout some 
time ago, which triggered me to look at those classifications closer.


Reading that one wants to classify a road depending on how it's uses (so 
what rolls over it and how much) is not correct imho.  You need to look 
at the designation of the road.   the N1 is an easy example of course.


And now we can start to discuss what big city is in Belgium ;-) But most 
of them are easy to classify as such with some common sense.   I would 
still say that primary roads are easy, there is much more difficulty in 
determining lower type of roads.


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


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Ben Laenen
On Thursday 27 December 2012 18:15:05 Jo wrote:
> Now that we're on the subject of road classification, The northern part of
> the ring of Leuven has separate lanes for both directions, no traffic
> lights, on and off ramps like a motorway and the maximum speed is 90 km/h
> (a rare occurence these days in Flanders).
> The southern part has crossings with traffic lights and a speed limit of 50
> km/h, complete with a truckload of traffic cams to enforce it.
> 
> At some point I had tagged the northern part as trunk, since it's  far more
> interesting to go that way from east to west or west to east, so why
> wouldn't we visualise that on a rendered map? Somebody retagged the whole
> ring road as primary afterwards and I left it as such, since I didn't feel
> like starting an edit war. It still feels like a missed chance to be the
> better map though.

Doesn't the northern part of the ringway have cycle lanes on each side? So far 
the "definition" of trunk is a road for fast traffic that disallows 
pedestrians and cyclists.


To also reply to the next email: the original definition was about the 
motorroad sign, but that isn't really workable and it has changed into the 
definition above. There are roads with the sign that shouldn't be trunk and 
roads without the sign but which should really be trunk (these roads then have 
traffic signs that just prohibit pedestrians and cyclists (and mopeds, horse 
riders). Or the A12 between Antwerp and Brussels, I don't think the middle 
lane has any traffic signs at all on the parts where it isn't a motorway.

Ben

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


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Jo
Darn, they seem to have forgotten to put F9 traffic signs! I did forget to
mention the bicycle lane is completely separate, although they do cross the
on/off ramps.

I thought N25 was still trunk between De Mol (Tiensesteenweg) and Blanden
(we call it De Expressweg), but that's not true anymore.

The F9 sign is not present either on the only bit of road which still is
mapped as trunk in Leuven. Our connection between the Kapucijnenvoer on the
ring road and E40/E314, the N264/ Boudewijnlaan. This one doesn't have
cycle paths, but no signs to explicitely prohibit cyclists from using it.
Hardly any cyclists consider using the escape lane though. There are better
alternatives over the university campus, which look a lot more inviting.

I guess I'll have to rest my case and put it to bed. Tuck it in and never
look back :-)

Jo

2012/12/27 Johan C 

> Sign F9 = autoweg = trunk
>
> http://nl.wikipedia.org/wiki/Verkeersborden_in_Belgi%C3%AB_-_Serie_F:_Aanwijzingsborden
>
> Nice to read on this topic (sorry, it's in |Dutch):
> http://forum.openstreetmap.org/viewtopic.php?id=16232
> http://forum.openstreetmap.org/viewtopic.php?id=16231
>
> Cheers, Johan
>
>
> 2012/12/27 Jo 
>
>> Now that we're on the subject of road classification, The northern part
>> of the ring of Leuven has separate lanes for both directions, no traffic
>> lights, on and off ramps like a motorway and the maximum speed is 90 km/h
>> (a rare occurence these days in Flanders).
>> The southern part has crossings with traffic lights and a speed limit of
>> 50 km/h, complete with a truckload of traffic cams to enforce it.
>>
>> At some point I had tagged the northern part as trunk, since it's  far
>> more interesting to go that way from east to west or west to east, so why
>> wouldn't we visualise that on a rendered map? Somebody retagged the whole
>> ring road as primary afterwards and I left it as such, since I didn't feel
>> like starting an edit war. It still feels like a missed chance to be the
>> better map though.
>>
>> Jo
>>
>> 2012/12/27 Kevin Grossard 
>>
>>>  > On Thursday 27 December 2012 09:51:36 Kevin Grossard wrote:
>>> > > The wiki about the highway conventions distinguishes primairy,
>>> secondary
>>> > > and tertiairy roads using the N-numbers (although there are some
>>> question
>>> > > marks). Using the current conventions means using the old
>>> classification
>>> > > when the N-numbers were given. Various N-ways got reconstructed, the
>>> > > traffic got redirected by other roads, some roads aren't suitable
>>> anymore
>>> > > for the current traffic.
>>> > >
>>> > > The spatial structure plan for Flanders (ruimtelijk structuurplan
>>> > > Vlaanderen) has a list with the primairy roads [pages 368-377
>>> > > http://www2.vlaanderen.be/ruimtelijk/docs/rsv2011/RSV2011.pdf]. I
>>> suggest
>>> > > to adjust the wiki so there will be an uniform highway classification
>>> > > where primairy roads are the roads selected in the structure plan.
>>> The
>>> > > other N-roads can be describes as secondary roads. Tertiary roads
>>> can be
>>> > > described as other important local roads (nl: steenwegen die geen
>>> > > secundaire weg zijn). Minor or residential roads can be understand as
>>> > > local roads (nl: lokale wegen). What do you think?
>>> >
>>> > I don't really think we can just take that classification and apply it
>>> to OSM.
>>> > It may not look like it from the map in that file, but it would make
>>> Flanders
>>> > almost void of primary roads. Our government isn't very keen to give
>>> lots of
>>> > roads a primary status. And all the dead-end primary roads won't give
>>> a nice
>>> > map either (ringway of Lier without any primary road connecting to it,
>>> things
>>> > like that).
>>> >
>>> > I've been aware about the official classification (read some previous
>>> > discussions in the mailing list archive), but I think we at least need
>>> to
>>> > include some secundary classified roads tagged as primary in OSM to
>>> make a
>>> > useful map. So far, we don't have a list of those. In the UK, the
>>> category of
>>> > official primary roads would be tagged as trunk in OSM.
>>> >
>>> > That said, using road numbers to determine OSM classification is
>>> actually how
>>> > it's done in most countries. It's not perfect by far but we've always
>>> allowed
>>> > some deviations from the rule where it makes sense.
>>> >
>>> > Ben
>>>
>>> Okay, there aren't a lot of primary roads in Belgium but that's the
>>> result of the historical urban planning and short term vision.
>>> Fact is that primary roads also differ in reality from the secondary
>>> roads. For example like the surface, the lack of houses, shops and
>>> schools (lineair settlement and crossings through villages), the lack of
>>> cycle tracks, ...
>>>
>>> Using the road numbers is a lot easier and that's a good argument but
>>> the goal of OSM isn't making the Belgium roads more attractive, i suppose?
>>> I don't get why a dead end primary road connected to a seconda

Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Johan C
Sign F9 = autoweg = trunk
http://nl.wikipedia.org/wiki/Verkeersborden_in_Belgi%C3%AB_-_Serie_F:_Aanwijzingsborden

Nice to read on this topic (sorry, it's in |Dutch):
http://forum.openstreetmap.org/viewtopic.php?id=16232
http://forum.openstreetmap.org/viewtopic.php?id=16231

Cheers, Johan


2012/12/27 Jo 

> Now that we're on the subject of road classification, The northern part of
> the ring of Leuven has separate lanes for both directions, no traffic
> lights, on and off ramps like a motorway and the maximum speed is 90 km/h
> (a rare occurence these days in Flanders).
> The southern part has crossings with traffic lights and a speed limit of
> 50 km/h, complete with a truckload of traffic cams to enforce it.
>
> At some point I had tagged the northern part as trunk, since it's  far
> more interesting to go that way from east to west or west to east, so why
> wouldn't we visualise that on a rendered map? Somebody retagged the whole
> ring road as primary afterwards and I left it as such, since I didn't feel
> like starting an edit war. It still feels like a missed chance to be the
> better map though.
>
> Jo
>
> 2012/12/27 Kevin Grossard 
>
>>  > On Thursday 27 December 2012 09:51:36 Kevin Grossard wrote:
>> > > The wiki about the highway conventions distinguishes primairy,
>> secondary
>> > > and tertiairy roads using the N-numbers (although there are some
>> question
>> > > marks). Using the current conventions means using the old
>> classification
>> > > when the N-numbers were given. Various N-ways got reconstructed, the
>> > > traffic got redirected by other roads, some roads aren't suitable
>> anymore
>> > > for the current traffic.
>> > >
>> > > The spatial structure plan for Flanders (ruimtelijk structuurplan
>> > > Vlaanderen) has a list with the primairy roads [pages 368-377
>> > > http://www2.vlaanderen.be/ruimtelijk/docs/rsv2011/RSV2011.pdf]. I
>> suggest
>> > > to adjust the wiki so there will be an uniform highway classification
>> > > where primairy roads are the roads selected in the structure plan. The
>> > > other N-roads can be describes as secondary roads. Tertiary roads can
>> be
>> > > described as other important local roads (nl: steenwegen die geen
>> > > secundaire weg zijn). Minor or residential roads can be understand as
>> > > local roads (nl: lokale wegen). What do you think?
>> >
>> > I don't really think we can just take that classification and apply it
>> to OSM.
>> > It may not look like it from the map in that file, but it would make
>> Flanders
>> > almost void of primary roads. Our government isn't very keen to give
>> lots of
>> > roads a primary status. And all the dead-end primary roads won't give a
>> nice
>> > map either (ringway of Lier without any primary road connecting to it,
>> things
>> > like that).
>> >
>> > I've been aware about the official classification (read some previous
>> > discussions in the mailing list archive), but I think we at least need
>> to
>> > include some secundary classified roads tagged as primary in OSM to
>> make a
>> > useful map. So far, we don't have a list of those. In the UK, the
>> category of
>> > official primary roads would be tagged as trunk in OSM.
>> >
>> > That said, using road numbers to determine OSM classification is
>> actually how
>> > it's done in most countries. It's not perfect by far but we've always
>> allowed
>> > some deviations from the rule where it makes sense.
>> >
>> > Ben
>>
>> Okay, there aren't a lot of primary roads in Belgium but that's the
>> result of the historical urban planning and short term vision.
>> Fact is that primary roads also differ in reality from the secondary
>> roads. For example like the surface, the lack of houses, shops and
>> schools (lineair settlement and crossings through villages), the lack of
>> cycle tracks, ...
>>
>> Using the road numbers is a lot easier and that's a good argument but the
>> goal of OSM isn't making the Belgium roads more attractive, i suppose? I
>> don't get why a dead end primary road connected to a secondary road can't
>> be useful. They will look stupid but that's reality and where
>> urban planning in Belgium is all about.
>>
>> But like you said, it would be useful to make a list with exceptions to
>> make and keep it simple if that's the way you want to keep it.
>>
>> Kevin
>>
>>
>> ___
>> Talk-be mailing list
>> Talk-be@openstreetmap.org
>> http://lists.openstreetmap.org/listinfo/talk-be
>>
>>
>
> ___
> Talk-be mailing list
> Talk-be@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-be
>
>
___
Talk-be mailing list
Talk-be@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-be


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Sander Deryckere
2012/12/27 Jo 

> Now that we're on the subject of road classification, The northern part of
> the ring of Leuven has separate lanes for both directions, no traffic
> lights, on and off ramps like a motorway and the maximum speed is 90 km/h
> (a rare occurence these days in Flanders).
> The southern part has crossings with traffic lights and a speed limit of
> 50 km/h, complete with a truckload of traffic cams to enforce it.
>
> At some point I had tagged the northern part as trunk, since it's  far
> more interesting to go that way from east to west or west to east, so why
> wouldn't we visualise that on a rendered map? Somebody retagged the whole
> ring road as primary afterwards and I left it as such, since I didn't feel
> like starting an edit war. It still feels like a missed chance to be the
> better map though.
>
> Jo
>
>
There has been some German discussion about what "trunk" means after all.
And it's relevant for Belgium too. As the "autoweg" sign has absolutely no
meaning. The maximum speed is the same as for normal roads (120 when there
are separate lanes, 90 default and 50 in residential area), so that sign
really has no meaning.

The (or at least some) Germans wanted to use "trunk" for the types of road
where slow traffic is forbidden (no agricultural traffic and no bicycles).
I think that's a far better definition than our current trunk definition.

Regards,
Sander
___
Talk-be mailing list
Talk-be@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-be


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Jo
Now that we're on the subject of road classification, The northern part of
the ring of Leuven has separate lanes for both directions, no traffic
lights, on and off ramps like a motorway and the maximum speed is 90 km/h
(a rare occurence these days in Flanders).
The southern part has crossings with traffic lights and a speed limit of 50
km/h, complete with a truckload of traffic cams to enforce it.

At some point I had tagged the northern part as trunk, since it's  far more
interesting to go that way from east to west or west to east, so why
wouldn't we visualise that on a rendered map? Somebody retagged the whole
ring road as primary afterwards and I left it as such, since I didn't feel
like starting an edit war. It still feels like a missed chance to be the
better map though.

Jo

2012/12/27 Kevin Grossard 

>  > On Thursday 27 December 2012 09:51:36 Kevin Grossard wrote:
> > > The wiki about the highway conventions distinguishes primairy,
> secondary
> > > and tertiairy roads using the N-numbers (although there are some
> question
> > > marks). Using the current conventions means using the old
> classification
> > > when the N-numbers were given. Various N-ways got reconstructed, the
> > > traffic got redirected by other roads, some roads aren't suitable
> anymore
> > > for the current traffic.
> > >
> > > The spatial structure plan for Flanders (ruimtelijk structuurplan
> > > Vlaanderen) has a list with the primairy roads [pages 368-377
> > > http://www2.vlaanderen.be/ruimtelijk/docs/rsv2011/RSV2011.pdf]. I
> suggest
> > > to adjust the wiki so there will be an uniform highway classification
> > > where primairy roads are the roads selected in the structure plan. The
> > > other N-roads can be describes as secondary roads. Tertiary roads can
> be
> > > described as other important local roads (nl: steenwegen die geen
> > > secundaire weg zijn). Minor or residential roads can be understand as
> > > local roads (nl: lokale wegen). What do you think?
> >
> > I don't really think we can just take that classification and apply it
> to OSM.
> > It may not look like it from the map in that file, but it would make
> Flanders
> > almost void of primary roads. Our government isn't very keen to give
> lots of
> > roads a primary status. And all the dead-end primary roads won't give a
> nice
> > map either (ringway of Lier without any primary road connecting to it,
> things
> > like that).
> >
> > I've been aware about the official classification (read some previous
> > discussions in the mailing list archive), but I think we at least need
> to
> > include some secundary classified roads tagged as primary in OSM to make
> a
> > useful map. So far, we don't have a list of those. In the UK, the
> category of
> > official primary roads would be tagged as trunk in OSM.
> >
> > That said, using road numbers to determine OSM classification is
> actually how
> > it's done in most countries. It's not perfect by far but we've always
> allowed
> > some deviations from the rule where it makes sense.
> >
> > Ben
>
> Okay, there aren't a lot of primary roads in Belgium but that's the result
> of the historical urban planning and short term vision.
> Fact is that primary roads also differ in reality from the secondary
> roads. For example like the surface, the lack of houses, shops and
> schools (lineair settlement and crossings through villages), the lack of
> cycle tracks, ...
>
> Using the road numbers is a lot easier and that's a good argument but the
> goal of OSM isn't making the Belgium roads more attractive, i suppose? I
> don't get why a dead end primary road connected to a secondary road can't
> be useful. They will look stupid but that's reality and where
> urban planning in Belgium is all about.
>
> But like you said, it would be useful to make a list with exceptions to
> make and keep it simple if that's the way you want to keep it.
>
> Kevin
>
>
> ___
> Talk-be mailing list
> Talk-be@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-be
>
>
___
Talk-be mailing list
Talk-be@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-be


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Kevin Grossard

> On Thursday 27 December 2012 09:51:36 Kevin Grossard wrote:
> > The wiki about the highway conventions distinguishes primairy, secondary
> > and tertiairy roads using the N-numbers (although there are some question
> > marks). Using the current conventions means using the old classification
> > when the N-numbers were given. Various N-ways got reconstructed, the
> > traffic got redirected by other roads, some roads aren't suitable anymore
> > for the current traffic.
> > 
> > The spatial structure plan for Flanders (ruimtelijk structuurplan
> > Vlaanderen) has a list with the primairy roads [pages 368-377
> > http://www2.vlaanderen.be/ruimtelijk/docs/rsv2011/RSV2011.pdf]. I suggest
> > to adjust the wiki so there will be an uniform highway classification
> > where primairy roads are the roads selected in the structure plan. The
> > other N-roads can be describes as secondary roads. Tertiary roads can be
> > described as other important local roads (nl: steenwegen die geen
> > secundaire weg zijn). Minor or residential roads can be understand as
> > local roads (nl: lokale wegen). What do you think?
> 
> I don't really think we can just take that classification and apply it to 
> OSM. 
> It may not look like it from the map in that file, but it would make Flanders 
> almost void of primary roads. Our government isn't very keen to give lots of 
> roads a primary status. And all the dead-end primary roads won't give a nice 
> map either (ringway of Lier without any primary road connecting to it, things 
> like that).
> 
> I've been aware about the official classification (read some previous 
> discussions in the mailing list archive), but I think we at least need to 
> include some secundary classified roads tagged as primary in OSM to make a 
> useful map. So far, we don't have a list of those. In the UK, the category of 
> official primary roads would be tagged as trunk in OSM.
> 
> That said, using road numbers to determine OSM classification is actually how 
> it's done in most countries. It's not perfect by far but we've always allowed 
> some deviations from the rule where it makes sense.
> 
> Ben

Okay, there aren't a lot of primary roads in Belgium but that's the result of 
the historical urban planning and short term vision.
Fact is that primary roads also differ in reality from the secondary roads. For 
example like the surface, the lack of houses, shops and schools (lineair 
settlement and crossings through villages), the lack of cycle tracks, ...
 
Using the road numbers is a lot easier and that's a good argument but the goal 
of OSM isn't making the Belgium roads more attractive, i suppose? I don't get 
why a dead end primary road connected to a secondary road can't be useful. They 
will look stupid but that's reality and where urban planning in Belgium is all 
about.
 
But like you said, it would be useful to make a list with exceptions to make 
and keep it simple if that's the way you want to keep it.
 
Kevin
  ___
Talk-be mailing list
Talk-be@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-be


Re: [OSM-talk-be] wiki highway conventions

2012-12-27 Thread Ben Laenen
On Thursday 27 December 2012 09:51:36 Kevin Grossard wrote:
> The wiki about the highway conventions distinguishes primairy, secondary
> and tertiairy roads using the N-numbers (although there are some question
> marks). Using the current conventions means using the old classification
> when the N-numbers were given. Various N-ways got reconstructed, the
> traffic got redirected by other roads, some roads aren't suitable anymore
> for the current traffic.
> 
> The spatial structure plan for Flanders (ruimtelijk structuurplan
> Vlaanderen) has a list with the primairy roads [pages 368-377
> http://www2.vlaanderen.be/ruimtelijk/docs/rsv2011/RSV2011.pdf]. I suggest
> to adjust the wiki so there will be an uniform highway classification
> where primairy roads are the roads selected in the structure plan. The
> other N-roads can be describes as secondary roads. Tertiary roads can be
> described as other important local roads (nl: steenwegen die geen
> secundaire weg zijn). Minor or residential roads can be understand as
> local roads (nl: lokale wegen). What do you think?

I don't really think we can just take that classification and apply it to OSM. 
It may not look like it from the map in that file, but it would make Flanders 
almost void of primary roads. Our government isn't very keen to give lots of 
roads a primary status. And all the dead-end primary roads won't give a nice 
map either (ringway of Lier without any primary road connecting to it, things 
like that).

I've been aware about the official classification (read some previous 
discussions in the mailing list archive), but I think we at least need to 
include some secundary classified roads tagged as primary in OSM to make a 
useful map. So far, we don't have a list of those. In the UK, the category of 
official primary roads would be tagged as trunk in OSM.

That said, using road numbers to determine OSM classification is actually how 
it's done in most countries. It's not perfect by far but we've always allowed 
some deviations from the rule where it makes sense.

Ben

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