Proposal:Highway=tertiary

The Feature Page for the approved proposal highway=tertiary is located at Tag:highway=tertiary

highway:tertiary
Proposal status:	Approved (active)
Proposed by:	Rw
Tagging:	highway=tertiary Data item
Applies to:	ways
Definition:	A class of road between unclassified and secondary
Statistics:
Rendered as:	*
Vote end:	2006-11-18

Originally proposed by Rw on 01:01, 25 October 2006.

Tertiary addresses two issues for me. Around here there are a large number of significant roads that are clearly more substantial than "unclassified" and less substantial than "secondary". In (sub)urban areas this might be the road that leads to the secondary road; it's mixed use rather than residential and perhaps two or three lanes plus flares at the intersection, but clearly not a secondary road itself.

As a suburban example I offer Harvest Hills Drive NE in Calgary, Alberta, Canada. Harvest Hills Drive NE (and Harvest Hills Gate NE) are residential, but they are distinguished by being the dominant road in the subdivision. All the other roads defer to these at intersections (stop and yield signs). These roads are wider and bear the bulk of the rush-hour traffic.

As a rural example, see every township road in Ontario. In Ontario, many highways are maintained by the province. These provincial highways equate nicely with motorway, primary and secondary. That leaves a large number of rural roads that are maintained by the local municipality / township with nothing to distinguish them from unclassified or track. These may be oiled gravel and clearly not secondary but are actually classified. They are classified as municipal or township or county roads and 80km/h is common.

Another important distinction that might not be apparent to observers in other places. When it snows, roads are generally cleared in order of importance. These tertiary roads will be cleared earlier and more often than the residential, industrial, service or other roads around them. And in this part of the world, that is significant for a portion of the year.

Tertiary is a natural extension to the hierarchy of primary, secondary, ... and is certain to apply in other jurisdictions. It is also a generic description that will allow OSM contributors to select it without having to match a location-specific criteria.

Discussion

I think this tag should be added as there is (at least in the UK) another class of road between secondary and unclassified. There are a number of roads around here that are not secondary roads but are far to big/fast to be labelled as unclassified, most of which turn out to be C roads (tertiary).

The OS render these roads as yellow but using a different thickness.

--Dean Earley 13:41, 5 November 2006 (UTC)

One problem is that this is a UK-centric scheme and it's really hard to find a rigorous adaptation in others country (France for example). What about the progression of the dual Map Features scheme (physical/administrative) ? FredB 13:47, 5 November 2006 (UTC)

Well, the naming is actually, 1st, 2nd, 3rd, etc.. which I'd have thought maps fairly easily onto pretty much every countrys scheme. I (and I assume some others in the uk) use Motorway, trunk, A, B, C, and minor. --Dean Earley 15:13, 5 November 2006 (UTC)

This feature was also requested by Wikiproject Ontario, Canada, so it is not UK-centric. Bruce89 18:50, 5 November 2006 (UTC)

Besides, surely it's better to have a certain amount of redundancy in the symbols/options available, rather than have too few. I don't see what the problem would be if it makes the map clearer in one country, but isn't used in another. Richard B 13:11, 10 November 2006 (UTC)

I'm mapping areas of countryside, and really need a way of distinguishing between fast, decent, but not-quite-B roads (pale orange on OS Explorer maps), and single track lanes where you'll have to reverse half a mile if someone's coming the other way (yellow). Unclassified for both doesn't cut it; tertiary for one would be spot on. southglos 19:10, 5 November 2006 (UTC)

I just creating prpousal for in-town highway value, little more you may found on Croatian roads taging, Dubravko 7:37, 6 November 2006 (UTC)

There is no way of finding if a road is a tertiary. Signs don't indicate it, and nore does anything else. It would be far more practical, to use the highway=unclassified tag for roads that are under x feet/meters wide, and then use the tertiary tag for roads that are beneath secondary, and more than that width. I think the tertiary should then be rendered as unclassified for now, and unclassified should be thinner, but the core and border colours shoudl remain the same. This would be more approprate for use in countries that arnt in the UK, and would mean the data is available to gather. Ben. 02:15 11/11/06 (UTC)

Agree that there's no easy way to find out what's officially a (UK) "C" road, but your comment on using "tertiary" as meaning wide-unclassified and "unclassified" as narrow-unclassified is, more or less how I intend to use it (although width isn't the only qualifier - you can get wide roads that aren't really suitable for through traffic). So, I wouldn't reject the proposal just on a strict "tertiary" must mean "C-road" basis; it's valuable (I'd say essential) for marking up "preferred" routes through a jumble of unclassified roads. --Southglos 11:54, 11 November 2006 (UTC)

I'm not in disagreement that there needs to be a way of defining road thicknesses. Travvelling on a smaller minor road can be half the speed of the larger ones. I'm for tagging the unclassifieds and the small unclassified lanes differently, I just feal this isn't the most logical way to do so. Just tagging the smaller ones as 'narrow' makes more sence to me.Ben. 12:15 11/11/06 (UTC)

A good criterion to distinguish tertiary and unclassified would be "is it comfortable to walk along without frequently encountering cars". If yes, it's unclassified. If not, it's tertiary. That way, if you plan to do a walk in the countryside you can use unclassified roads but avoid tertiary roads if possible Nick W 09:01 12/11/06 (UTC)

That doesnt really work, as round me there are some roads that are relitively wide as they are built to carry quite a bit of traffic for events, but for the rest of the year they have next to no traffic. That would mean that these large minor roads, are tagged as unclassified. Some small roads are used as shortcuts also. This would mean they appear as a Main minor road, while infact they are really unclassified (judging them by there width). OS make the 2 different thicknesses of render based on the thickness of the road. (normal minors, and minors less that 4M in width) This is not only a good graphic represeentation, but also corelatees perfectly to the potential capacity of the road, and does not demand data that is unavalibale for roads to be added. Why not learn from them, and tag the roads width if nessesry. Although a tag is very much needed, I'm not going to be able to use this current proposal at present, without guessing heavily, so nomatter what the outcome, its worthless. Ben. 13:11 12/11/2006 (UTC)

Rendering

I use the following ruls in my OSM map features file for tertiary road:

<rule e="segment|way" k="highway" v="tertiary">
	<line class='highway-tertiary-casing' />
</rule>
               
<rule e="segment|way" k="highway" v="tertiary">
	<rule k="oneway" v="~">
		<line class='highway-tertiary-core' />	
	</rule>
	<rule k="oneway" v="1|yes|true">
		<line class='highway-tertiary-core oneway' />	
	</rule>		
	<rule k="oneway" v="-1">
		<line class='highway-tertiary-core otherway' />		
	</rule>		
</rule>

Inside the rendername section:

<rule e="segment|way" k="highway" v="tertiary">
	<text k="name" text-anchor='middle' startOffset='50%' class="highway-tertiary-name" />
</rule>

And renderref:

<rule e="segment|way" k="highway" v="tertiary">
	<text k="ref" class='highway-tertiary-ref' dx='2.5px' dy='-2.5px' />
</rule>

And finally, the styles:

.highway-tertiary-casing {
  stroke-width: 2px; 
  stroke-linecap: butt; 
  stroke-linejoin: round;
  stroke: #222222;
  fill: none;
  }

.highway-tertiary-core {
  stroke-width: 1.5px; 
  stroke-linecap: butt; 
  stroke-linejoin: round;
  stroke: #FEF970;
  fill: none;
  }

.highway-tertiary-name {
  fill: black;
  font-family: verdana;
  font-size: 1px;
  font-weight: bolder;
  baseline-shift: -35%;
  }

.highway-tertiary-ref {
  fill: black;
  stroke: white;
  stroke-width: 0.3px;
  font-family: verdana;
  font-size: 5px;
  font-weight: bolder;
  }			

Votes

Open for Voting until the 18th of November 2006

• I approve this proposal. Dean Earley 00:40, 11 November 2006 (UTC)
• I approve this proposal. Gagravarr 01:25, 11 November 2006 (UTC)
• I partly disaprove of this proposal, for the reason above. Data can't come from nowhere...So if this goes ahead as it currently is then I won't be able to use this tag. I approve that the different roads must be taged seperate though.Ben. 02:12, 11 November 2006 (UTC)
• I approve this proposal. --Southglos 11:55, 11 November 2006 (UTC)
• I approve this proposal. --Dubravko 12:51, 11 November 2006 (UTC)
• I approve this proposal, there are lots of roads like this in Scotland. Bruce89 14:36, 11 November 2006 (UTC)
• I approve this proposal. TomChance 16:05, 11 November 2006 (UTC)
• I approve this proposal. Rw 17:17, 11 November 2006 (UTC)
• I approve this proposal. Nick W 09:03, 12 November 2006 (UTC)
• I approve this proposal. Matthew Newton 02:08, 14 November 2006 (UTC)
• I approve this proposal. --EdoM 10:58, 14 November 2006 (UTC)
• I approve this proposal. RalfZ 20:49, 18 November 2006 (UTC)

Results

So highway=tertiary has been accepted but there seems to be some confusion over whether this is strictly C class roads (in the UK) which are almost impossible to identify accurately or whether we use some other scheme like ">4m wide" or "has white lines down the centre". Please note that the local councils most likely use exactly these distinctions to signify C roads.

How long before this is built into osm-map-features.xml? I'm not really confident in editing the file myself for fear of borking it. 17:21, 21 November 2006 (UTC)

It is now supported by osmarender in the same way as highway=minor, which this in-effect replaces. Gagravarr 18:11, 21 November 2006 (UTC)