Instead of tagging addr:postcode with either 11111 (ZIP), 11111-1111 (ZIP+4), or 11111;11111-1111 (ZIP;ZIP+4), is there something like addr:postcode:plus_four? ZIP+4 codes can change and so I don’t want to only tag the +4 as it could go out of date quickly.
We’ve been putting ZIP+4 in addr:postcode=* when known. This includes both manual entry and bulk address imports.
Do you have any idea how unstable the extension can be? I know the five-digit ZIP code can only change at most once a decade, and only in exceptional circumstances, but I don’t know much about how the extension is assigned. I’ve personally never experienced a change in extension while remaining at an address for a couple decades across many changes in routes, but maybe some other kinds of routes experience more churn?
Should that really be a + sign in the ZIP+ code? I was under the impression that the convention is to tag it as addr:postcode=11111-1111.
The written convention is always a hyphen and that’s how we do it in OSM too. The plus sign is only in the name of the system.
What I’ve seen is that the +4 code helps USPS by specifying a route which helps them. The routes aren’t as un changeable as zip codes are because they’re routes which USPS may change.
The possibility of churn is good to keep in mind; maybe we can document it somewhere just as a heads-up. I wouldn’t worry too much about that part of addr:postcode=* getting out of date. We tag plenty of other stuff that’s less stable anyways, like website=*, opening_hours=*, addr:unit=*, and shops themselves.
Wouldn’t it make more sense to have another key for that addition?
No, because it’s a single number that always goes together. You’d never use the +4 extension on its own. The five-digit ZIP code also consists of a few components, such as the delivery area, that we wouldn’t break out separately.
Though I see here a different situation. The first 5 digits are commonly known and can be verified quite easily. Like shops will have those in their address printed on the door…
But the 4 additional digits you mainly can get from USPS database.
So how would someone be able to express he verified the postal code is correct without having a clue about the addition?
As of last month, 123,600 distinct values of addr:postcode=#####-#### occur in the U.S. How did we come up with so many ZIP+4 codes, and how do you verify that one of these codes is correct? Exactly the same as with any five-digit ZIP code. Most came in through imports of municipal or county address datasets. Many more are the result of a mapper observing a ZIP+4 on the POI’s sign or website, a receipt that came with a purchase, or a flyer that came in the mail. It’s an extra level of detail that any software capable of handling ZIP codes would need to handle anyways, whether in our database or in user input.
To verify any ZIP or ZIP+4 that we have in the database, you can either go back to these original sources or look up the address on the USPS website. Don’t copy addresses from the USPS site, but verifying them there should be fine.