[Talk-us] TIGER 2022 PLACE dataset

[Brian M. Sperlongano]

If it's purely redundant (same boundaries and name) then that seems silly
to keep around.

There are some cases where the CDP is a subset of a larger municipal
boundary.  For example, the Wakefield-Peacedale CDP in Rhode Island,
encloses two villages that are so close to each other that they're often
spoken of in tandem.  However, both are entirely enclosed by the town of
South Kingstown which encompasses a much larger area.

In my use case, I don't consume Wakefield-Peacedale in my software.

Is it useless?  I don't know.  Does it fill a niche not covered by a
municipal boundary?  No.

CDP: https://www.census.gov/quickfacts/wakefieldpeacedalecdprhodeisland
CDP in OSM: https://www.openstreetmap.org/relation/191222
South Kingstown admin boundary:
https://www.openstreetmap.org/relation/7539760

On Sat, Jan 21, 2023 at 12:36 PM Zeke Farwell <ezekielf at gmail.com> wrote:

> Ok, so what I'm hearing is that boundary=census (CDP) is a useful boundary
> type to include in OSM when a named populated place is unincorporated and
> therefore has no administrative boundary to include in OSM.  Including
> boundary=census is useful in these cases so data consumers working with
> municipal level boundaries have a fallback to use where no administrative
> boundary exists.  That being the case, it sounds like in a place where a
> municipal level admin boundary does exist, a boundary=census serves no
> purpose?  So in most of New England where all the land is sliced up into
> municipalities we can feel free to delete any boundary=census objects where
> a municipal admin boundary with the same name exists?
>
> _______________________________________________
> Talk-us mailing list
> Talk-us at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-us/attachments/20230121/0afbd037/attachment.htm>