Don’t do that! :shock:
If anything else goes into this mix, something is going to spontaneously combust or even explode - quite possibly my brain. 
There’s one aspect of Iain’s suggestion that worries me, but I don’t know enough about how Waze works to know whether it would be a real problem.
If all the major highways have all their segments populated with the post-town, how will that be affected by having potentially different local names in the minor highways, primary streets and streets, nearby?
I presume the A21 would go from being labelled Catford in SE6 to Bromley when it enters BR1. If you were to have Bellingham and Downham named on other streets nearby as the A21 passes through those localities, could Waze start reporting problems on the A21 between Catford and Bellingham, or would it be limited to the A21 in Catford?
I suspect it might cause some confusion as I assume that Waze will use the name from whichever is the nearest polygon, eg the A282 near Stone scenario.Can two city polygons overlap and what’s the behaviour when that happens, or don’t they at all?
I’d like to propose that minor highways, primary streets and streets should not have a city named at all, except away from major highways. To provide more local detail in the suburbs, where I’d accept a locality name would be useful, it should limited to the streets occupied by the main retail area for that locality. So although Bellingham and Downham wouldn’t be used, Beckenham and Grove Park would be, but only on the segments where there is a high level of retail activity. When people set off on a journey, they have in mind a specific address or a retail centre.
For editing purposes, this can be checked by those with local knowledge or using Google or Bing maps to identify the streets with shops where there are no local editors.
So someone using the A222 between Beckenham and Penge would be alerted to delays in Beckenham, in Penge, or between the two depending on where they are. Sufficiently useful to plot an instant detour, without causing potential confusion by using locality names on the major highways.
I’ve no idea whether that adds to the discussion or confuses it further :? . It’s meant to apply only to London but I guess the principle would be applicable to all cities and large towns.
Fully agree, I think this should be a long term goal - as we know waze don’t do short term, in the meantime, there may be places all over that may need adjusting to work around current flaws.
Mostly definitely! The “polygons” (such as they are) are only created by Waze by drawing a boundary around each mention of the city. Unless people were very careful to only use one name in an area, a real mess occurs. Hence London was everywhere!
So having A21 Bromley running through a Downham polygon could cause confusion. My proposal for no cities on minors and streets near majors stands.
Sent from my iPhone using Tapatalk
Not sure it would cause a confusion at least for traffic reports on the major road. They would get their “correct” city name as set for that segment.
On the other hand, reports for segments with no city would get something like “near”, “in the neighborhood (sic)”, or “between”. I presume that Waze uses the nearest centroid of a city name for those.
I’m pretty sure the answer is “not at all”. Since we put the city names onto the South Circular segments, I have not seen one single traffic report of “A205 S Circular near …” or “A205 S Circular between …”.
As far as I can tell, you will only get “near” or “between” if the segment does not have it’s own city name set.
:evil: :lol: How about we put the city as “the neighbourhood of city” so the reports will say traffic in the neighbourhood of city :lol:
Sounds as though there’s a tacit acceptance of Iain’s proposals for major highway segments’ city field being populated with the post-town, or am I taking that too far?
What about minors/streets? Locality everywhere with editorial disagreements over boundaries, or just in the retail centres?
Sent from my iPhone using Tapatalk
I originally suggested only locating small areas - mainly to reduce the workload for the editors. However, I think David has pointed out that, if we leave segments without city names, someone will come along and fill them in - possibly incorrectly. If they are all filled in, then someone coming along to add a segment can use the neighbouring segments as a guide.
So I now tend towards having them all named.
Also, now I am aware of the tools available, it’s not such a chore to fill in the city name over a whole area.
Up until now, I’ve only seen the Wikipedia list for London post codes. I just had a quick search to see if there was a similar resource for Bromley (BRxx), Dartford (DAxx) and so on. And there is: This page lists the main postcode areas, with links to a breakdown for each area.
For example, YO goes to this page, which lists 31 YO codes covering 10 different post towns - and it’s the post towns I’m proposing to use in London.
Now 18 :o of those codes are just “YORK”. As the local man, you would have to judge whether that was too large an area - and this could still happen with some of the outlying London codes - I haven’t had a good look. If so, we might have to produce our own “approve mapping” of post codes to town names and put it on the Wiki for reference. As long as we can agree on something, we can then try and stick to it.
Fair enough. The Wiki will need updating with the new rules, and we’ll need to set up a Boundaries Commission to arbitrate disputes 
Tools?
Sent from my GT-N7100 using Tapatalk 2
Is that what you think of upstart new editors who want to delete entire cities?
Couldn’t possibly comment!
Sent from my GT-N7100 using Tapatalk 2