Proposed import of All the Places data onto existing big box retail POIs in the US

I’m wrapping up the import of fast food and cafe data and am planning to move onto big box stores in the US. By “big box” stores, I basically chose brands that usually use a large footprint and are important anchors in a given retail zone. In other words, it’s a grab-bag of strip mall staples. The process will be the same as it was for hotels, grocery stores, and fast food: I will merely be adding tags to existing objects, not overwriting existing tags. I’ll continue to put nodes that do not match to an existing OSM objects in this cooperative MapRoulette challenge to be added manually and separately. More details on this stage of the import and links to the geojson files are on the OSM wiki page:

https://wiki.openstreetmap.org/wiki/Import/All_the_Places_US_data#Big_Box_Stores

Thank you to everyone who has left changeset comments or had other feedback on the work I’ve done so far! I think this might be the last round of these ATP imports I do. As always, I’m happy to answer any questions or field feedback!

2 Likes

I am preparing similar import and I discovered that comparing existing OSM data and ATP data where both have some tags present and investigating mismatches is a great way to find broken ATP spiders (and broken OSM data).

I reported so far multiple broken spiders, some were fixed already.

I expect that US data also has this type of problems and it would be worth fixing them before import.

3 Likes

Does All The Places also track store amenities that we tend to map separately, such as Costco Gasoline (outside Costco), CVS (often inside Target), and Banfield Pet Hospital (usually inside Petsmart)? It would be nice to take care of them together if possible, with the understanding that store hours and numbers may differ.

Does this list of chains include other store formats? For example, the name suggestion index has separate entries for Walmart versus Walmart Neighborhood Market.

1 Like

The answer to both of these questions is no. There are no gas station nodes mixed in with the various brands in this category (I’ll note this was fairly common back with the grocery store stage of the import). On the brand locations within other brands’ locations, ATP usually tracks these in the separate spiders as far as I can tell, including for all of the brands in this “big box” category. That includes the CVSs, Banfields, Starbucks, Ultas, etc that are often found inside Target, Walmart, Petsmart, etc locations.

1 Like

And to be more specific I am looking further into ATP data and started simple validations of email and phone:

from phone number validation ( validate phone numbers · Issue #9120 · alltheplaces/alltheplaces · GitHub ):

deleting phone = www.papersource.com as invalid
deleting phone = 011-555-290-6135 as invalid
deleting phone = 11-52-33-3127-0011 as invalid
deleting phone = 8.62134E+11 as invalid
deleting phone = () - as invalid
deleting phone = + as invalid
deleting phone = 89 as invalid
deleting phone = 95/7476120 509680523 as invalid
deleting phone = 09708 – 5909860 as invalid

from email validation ( require email to contain @ · Issue #9126 · alltheplaces/alltheplaces · GitHub )

deleting email = otwock.galeriakupiecka as invalid
deleting email = Konrad Pawlik as invalid
deleting email = 512-467-7041 as invalid

these examples are not limited to USA, but I expect that USA has at least some such cases