UNIVERSITY LA is recognized (if deprecated) as a place name for LSU by the USPS??? Do people write to GOVERNMENT DC, BOARDWALK NJ, BEACH HI, or SCHOOL CA?

OK maybe relying on USPS recognition to rule out silly city names isn't such a good idea.

I wanted to post another example of how relying on CDPs for primary naming can be problematic. Not only is there the case of "the City of UC Davis" I mentioned earlier, but California also has two CDPs named "Live Oak". One is Live Oak, a neighborhood of Santa Cruz that the Waze map helpfully calls "Live Oak CDP"; the other is Live Oak, an incorporated town north of Yuba City that the Waze map started out calling "Live Oak (2)". If we harmonized the place names to both be Live Oak we'd have a massive smudge, as they are roughly 150 miles apart.

(Separate issue is the number of years it takes for Waze to realize that one of its original cities needs its name changed; the old stuff seems to hang around in the back end in perpetuity.)

Perfect. Gives authority to local leadership while still providing some basic guidelines on analysis, and not too wordy or complicated. Two thumbs up.

ottonomy wrote:<snip>
I'm tempted to say that the only rule of Cities is that there are no rules, but that won't do.

My personal rule has always been "What city does the owner of the address use to tell people how to get to their property?" In many cases, it's quite easy to figure out, since there's a website for the business listing their address, or GIS parcel data for a residential area. Unfortunately, GIS is not universal, and has been demonstrated by several examples, probably biased. So, I'm with Otto, it's going to be a tough journey, but let's see what we can come up with.

I would propose that the Road names infobox text be relocated into the body of the section. It is not our typical Waze wiki style to pull guidance out into an infobox, and it reads as end-all guidance, whereas the later text clarifies it is not:

section body text wrote:There are other standards in place for other purposes requiring adding other street data in the the alternate. The addition of the USPS data does not change any other standards but is in addition to anything existing already on those affected segments.

Kartographer: when you transpose to the wiki, a grammatical error snuck into both places as "That may means many addresses...".

Kartografer wrote:

herrchin wrote: Kartographer: when you transpose to the wiki, a grammatical error snuck into both places as "That may means many addresses...".

I struck through the s when I first posted this, but it's hard to see.

Kartografer wrote:And everyone here seems to think my name has ph in it

I'll just... go home. I can't work the internet today. Sorry!

ehepner1977 wrote:

ottonomy wrote:<snip>
I'm tempted to say that the only rule of Cities is that there are no rules, but that won't do.

My personal rule has always been "What city does the owner of the address use to tell people how to get to their property?" In many cases, it's quite easy to figure out, since there's a website for the business listing their address, or GIS parcel data for a residential area. Unfortunately, GIS is not universal, and has been demonstrated by several examples, probably biased. So, I'm with Otto, it's going to be a tough journey, but let's see what we can come up with.

That was the basis of my argument in Discord. Waze is trying to get someone to an address almost always based on user-input from how that place is advertised (advertisement being me telling you my address over the phone, a web site, Facebook entry, etc.). It could be any manner of definitions, including colloquial names that have absolutely zero government recognition.

Many times they align with USPS primary, but not always, which is why the USPS may have several alternate names. Maybe it'll align with GIS, but not necessarily, as local plats are usually legally defined using a completely separate system (my house is NeighbornoodName 7th Addition, Block 5, Lot 17, and an ID value of 13 digits), and the street address in GIS could be barfed up from any number of sources.

All, please consider this post to be a placeholder for later (this evening? tomorrow?) when I have a chance to flesh out my thoughts below with explanation/background/evidence...

I am very pleased that this subject has come to the fore. I think it was grievously in need of work. But, I have some serious issues with the (natural) desires of those crafting the guidance to settle on one data source or another as more accurate, relevant, useful, reliable, etc.

USPS is rock solid in some places, and useless garbage in others.
CDPs are spot on in some places, but politically fabricated nonsense in others.

By "places" above, I do note mean entire states, or even entire metro areas. I can give examples inside of the Los Angeles area where one or the other source is preferable or perfect, and other examples where that same source is laughably wrong. We need standards by which we can objectively determine the rightness or wrongness. I fear that those standards might be excruciatingly difficult to produce in a reader friendly way, but I'm willing to join you all in the challenge.

I will also be making the argument that in some very specific cases, "neighborhoods" must be mapped, as they are the only names recognized by locals and visitors alike. Too specific for mention in the national wiki? Perhaps, but there needs to be some wiggle room there, or those who interpret every word of the wiki as a literal absolute will make life a continuous hell in those areas where rules must be broken.

I'm tempted to say that the only rule of Cities is that there are no rules, but that won't do.

Virginia has adopted a guideline that if the USPS name for addresses on a given segment is different from the primary city name on that segment (including "No city" as a primary name), then the USPS name should be added as an alternate name.

Given that we're unlikely to reach any nationwide consensus on primary city names due to state-to-state differences in how local governments are organized, how about adopting the Virginia guideline instead? A segment can have multiple alternate names, corresponding to variations on the city name accepted by USPS, so that's not a problem. If one of them isn't commonly used and no one ever searches for it, there's no harm done. But if a user does search for "123 Jones St, University, LA" for some reason, we'd have it on the map as an alternate name and the search would go to the correct location.

Note that the existing city name guidance only covers segments with "No city" as the primary name, so it doesn't cover situations like Queens where the primary USPS "city" name is not the same as the municipality (in this case, borough) name.

Kartografer wrote:One more thing:

DwarfLord wrote:there are plenty of unincorporated towns already shown on the Waze map with boundaries established by the census bureau that may already be somewhat arbitrary, and there is no reason for us to be chained strictly to whatever the bureau thinks

I'm assuming that you refer to CDPs here (those seem to be the imported unincorporated areas). CDP boundaries are arbitrary like FC is arbitrary, FWIW. Someone in the government, who is paid to determine these things, applied detailed criteria and made a decision. If CDPs are useful in general in one's state, their boundaries are worth following, rather than what we might think a visiting driver might think is a city.

Hear, hear.

CDP boundaries, like FC, are set by the government based on defined and definite criteria. While their boundaries might change every 10 years, FC maps often change more often than that. And despite initial skepticism, we can all agree now that FC is working.

A big part of the reason we implemented FC is that it takes the ambiguity out of the question entirely. Entire conversations and backs-and-forths about whether or not something should be Major or Minor no longer need to be had. The same should be true of city limits. Something is either in a city/CDP or it isn't.

The importance of CDPs vs. townships vs. whatever else vary from state to state and region to region—that's just a fact. There is no way to set a standard that applies to the entire country because different parts of the country have different laws, rules, and customs. But we should be able to agree on at least one thing: don't make it up. Set a standard and follow it.

I won't say exceptions can't ever be made, but that should be entirely a local decision.