Howdy neighbor. 
I grew up in Summit, GO HILLTOPPERS!!!, and I never knew that Short Hills was an unincorporated portion of Milburn. :shock: :shock:
I’ve always known it as Short Hills. Hell, the mall is called "The Mall at Short Hills (aka Short Hills Mall) and the address is: 1200 Morris Turnpike, Short Hills, NJ 07078
I just presented 4 examples, isn’t this an error?
Good points MW and Fredo. I think we’ve all determined that new guidance needs to be created.PS and orbit are determining what the appropriate hierarchy is so it’s clear what needs to change. This is statewide issue that needs to be addressed.
The State of NJ has a 73 page document of all local names they have on file, so it would be hasty to embark on this without having clear guidance.
Kobe is correct, there are no unincorporated places in NJ. Short Hills is not an incorporated municipality, but it is part of the incorporated municipality of Millburn.
However, for our purposes, yes, there are legitimate “city names” in NJ that are the names of unincorporated communities.
Note that some cities-within-cities are actually incorporated. For example, Metuchen is similar to Short Hills in that it is a smallish community within a larger city (Edison), but it is different form Short Hills in that it is separately incorporated, with its own elected government.
Ive been gone for a minute, now im back.
Nope, not an error.
I believe qwaletee addressed the main points already, so here’s for the sake of clarification. The word “place” can be broad, so for simplicity lets define “place” simply as “land” in this post. The US Census Bureau defines an Incorporated Place as “A type of governmental unit incorporated under state law”. The various types of government units that apply here, vary state by state in countless ways. (Thus making it impossible to set specific guidelines nationwide. Sorry PS and PDF.). NJ uses Counties and municipalities to define government units. The neighborhoods you mentioned are not considered unincorporated places, as the LAND falls under a municipality.
Qoute:
“Local government in New Jersey is composed of counties and municipalities. Local jurisdictions are simplier in New Jersey than in other states because, all of New Jersey is included in one of the state’s 21 counties and part of the state’s 565 municipalities. That means there are no independent cities or consolidated city-counties in New Jersey. There is no unincorporated territory. Every municipality is in exactly one county and every county has more than one municipality.”
Um, what fire? 
As of 2010 (possibly 2000), CDP’s are statistically equivalent to incorporated places, with the exceptions of not having hard borders and its own local government. (Although some states/countries will have both.) Also, in a recent Federal registrar (interesting read) I found:
Sounds awfully familiar to our mission
Isn’t Metuchen an incorporated borough surrounded by Edison Twp? The rule is, you cannot have a CDP in an officially incorporated city/place. Although this does not and cannot apply to the townships in NJ, because Townships = CDPs here.This issue was acknowledged by the Census Bureau 'sfar as I know.
Orbit, the Census Bureau addressed this topic giving every place its own unique code with added code for county-subdivisions (CDP,MCD,CCD)… Being that we already have segment IDs which hold magnitudes of data, wouldn’t it be possible to include the name of the CDP+County to each segment in that neighborhood? We can define the borders if they change (once every ten years?!) and everything can be handled in the back-end so as not to have conflicting data. Rules for city naming will also be clear-cut this way - either the county or local name.
Kobe,
Yes, Metuchen is a separately-incorporated borough within the city of Edison.
I dont think i understand your point of adding CDP+County. Could you explain the goal?
I think this would be the cleanest way to store the data, creating a new database row to store the CDP/County/whatever segment. As Kobe noted, this should be relatively static until the next census, but do know this would require a the servers to be updated which will obviously take some time.
The back end solution would be the cleanest.
populating dbs, creating associative views on a decade basis with minor changes at those increments should be relatively painless.
On another note, using primary/secondary naming conventions is fine for streets but is not available to places (points/areas) or PLRs, so granular searches will yield inherent error due to lack of available data, yes?
I don’t think any of this will be happening. If Waze had better search and didn’t use Google, it might be worth the effort, but as it stands, what are we hoping to gain?
Most simply, when trying to determine the name of a place, think, for any given road segment, if you were on it and asked a local “Where am I?”, what would they tell you? That’s your answer.
In past conversations, we established that the polygons generated by primary city names had meaning in their association with incorporated municipal names. That either turns out to not be the case, or at the very least is no longer the case, both of which are fine.
Anyway, the point is that a sub-city field may or may not be a solution to this problem. It may be the case that simply using CDP names as primary cities and entering municipal names (and postal names, if divergent) as alternate names turns out to be plenty sufficient. It’s just not clear at the moment, and the one thing I certainly want to avoid is setting a guideline now, causing a rush on mass city name changing, only to possibly have to do it again in the near future.
In the meantime, let’s worry about city names only on the immediate segments that need fixing to solve URs, and let’s stick to the current city-naming guidelines to resolve that - incorporated name in the primary name, and add any other names like CDP or postal cities, if different, to the alternate names so they get matched in searches. It’s not perfect, but it should be enough to solve URs while we get a more permanent solution figured out.
BTW, Metuchen is totally an incorporated borough, though completely surrounded by Edison township. Nearby Iselin and Colonia, however, are unincorporated towns within Woodbridge township.
The goal would be as minionsweb and pleasedrivefast said to resolve the duplicate place names. I actually felt this may be simpler than adding an entire new level for sub-cities. The more computer savvy among us may have better insight on this. Also, When you try to add a city name that is X distance away from a city with the same name, you get an error. Meaning the city names are already in the database somewhere not allowing duplicates. Adding a layer within the city name will accomplish the same thing as the addition to the segment ID. I completely agree that this conversation, while fun, gains absolutely nothing until the Waze navigation page is displayed first on the client.
Longer than creating a Sub city layer?
For PLRs you do have the alt field. For Places, good question. Being that these are POIs, using the local name would be right thing to do.
Phantom! Nice to see you come around! Having you playing devils advocate was no fun. Now that we are all (more or less) on the page, I just want to point out the original suggestion that we change the guidance to include CDPs.
I understand the concern of changing guidance only to be changed back later. My suggestion was not for editors to go on a mass editing spree and start including every CDP they find. in fact if there are no URs there should be no reason to change anything, as you stated. My concern here was the opposite extreme, of having CDP names from the base import DELETED or moved to he alt field en masse, again when there are no UR’s.
>The wiki FOR NOW can say simply “Incorporated places and CDP’s are used in the Primary Field. No changes should be made to existing CDP names unless they are blank or there are routing issues (ie. URs) involved. In either of those cases contact the SM for guidance.”
100% agree. It’s part of my obsession over inconsistencies in the way CDPs are treated, and knowing we had the FC update nearing completion, I wanted to at least resurrect this discussion.
Stay the course!
OK, Kobe, I see your point… for issues where we have a duplicate name that we need a better way to resolve, we definitely have options this way.
Where you can’t add the “correct” name because it is “too far,” there are several possibilities:
Waze decided to assign some incorrect city name to all the area between the segment you are changing and the closest like-named city. All those segments would need to be renamed, because they are wrong.
There is a duplicate city name, and the area you are editing “lost” and got named something a bit different. Follow the convention used for the surrounding area. If you think it is wrong, bring it up.
You made a mistake.
Not so likely in New Jersey, but in sparsely-developed areas, it is possible that a new set of roads will be added within a “city” limit that are far from any existing segments with that name. To resolve it, you can add a temporary segment or polygon at the edge of the existing defined area, then stretch it out to where you are. Waze will then see the city as extending to where you need it. I don’t recall for sure whether the “stretch” has an immediate effect, or you have to wait for some sort of update, but I think I was able to do this fairly quickly when I tried it. Don’t forget to delete your temp segment when you are done.
If we have conflicting CDPs, perhaps we should use the township name to segregate them?
Example:
Middletown township in Monmouth County has a CDP called Fairview. However, there is also an incorporated borough called Fairview in Bergen County.
Conflict Resolution:
Borough: Fairview
CDP: Fairview, Middletown Twp
It is 2 names in the same field, but I don’t know how else to both distinguish the 2 Fairviews in NJ and show the correct information on both the map and in search results.
Or you could just lift the PA standard en-masse and be done with it.
Do you have a link to the PA standard?
Here you go. Duplicate Cities.
This sounds good for most segments, but I’m still torn on what gets the primary city name for a road segment within a CDP whose name is different from both the local municipality name (usually township name) and the accepted city names for postal addresses.
In this situation, I don’t see any major problems arising from relegating the township name into one of the alternate name rows. However, the other 2 names are less clear. Colloquially, people probably identify with the CDP name, since that’s likely the reason the CDP exists. However, when searching for an address and reviewing any results returned to select from, it’s not unreasonable for people to enter and expect to see the postal city name.
Case Study: Fairton, Cumberland County. A large section of Fairfield township here goes by the CDP name, Fairton, but the accepted city name for all addresses within the township is actually Bridgeton, a nearby city. Between Bridgeton and Fairton, which name is more correct to display on the map and in search results?
Another Case Study: Princeton. The 08540 zip code (USPS-accepted city name: Princeton) spans parts of 5 different townships in 3 different counties with various CDPs within it. Which primary city name is correct? Princeton, reflecting the addresses for all those roads? Or the colloquial CDPs? The township names, since they are actually in different counties?
I don’t see NJ adopting the countyname scheme.
Agreed. 99.9% (if not 100%) of the time there are duplicate township names, either all or all but 1 commonly go by some other name, typically either accepted postal city names since those are most commonly reflected in addresses, or local CDP names.