Both, queens and non-queens addresses present

[ingalls]

Download the area here.
http://www.openstreetmap.org/browse/node/2555442506

For some reason these houses have an address such as ###-## and then one right next to it like ###.

Not sure what is going on, any advice? It is just this street, the rest of the import area looks good.

[colinreilly]

Yes, there are streets in NYC with both hyphenated and non-hyphenated addresses on the same block.

On this street, each house has two addresses in the NYC address point table. One most probably representing an historical address and the other current. This may well be an example of addresses being changed to comply with the standard throughout most of Queens with the exception of one outlier. 263 seems to be the only non-hyphenated address on this block. All others conform to the format 101-##.

[lxbarth]

On this street, each house has two addresses in the NYC address point table. One most probably representing an historical address and the other current. This may well be an example of addresses being changed to comply with the standard throughout most of Queens with the exception of one outlier.

@colinreilly - so would it be safe to say that if there is a non-hyphenated address and a hyphenated address in Queens, the non-hyphenated addresses is the historical one as Queens is moving towards Queens style addresses?

[colinreilly]

In general, yes but there are exceptions. In the Rockaways it is the reverse.

[lxbarth]

This is what I'm finding in the data. There seem to be areas where there's ongoing readdressing and other areas where the coexistence of hyphened and unhyphened addresses seems to be expected.

@colinreilly - I would love some guidance on how to resolve these address conflicts.

2339 instances where a building has a queens and a non-queens address: http://cl.ly/0G0z2m1u0e3R

[colinreilly]

Once again, the majority of these addresses are not incorrect. There is clearly a difference in approach to how the data is modeled and used. The NYC address points have multiple addresses at buildings for a variety of reasons but ultimately to ensure we get a 'hit' no matter how an address is referenced.

In the source data there is a special condition column (aka special address file [SAF] in City Planning parlance). If the value is not null, then the address is not your run-of-the-mill address.

A thorough definition for each value can be found starting on page 12 of the below reference pdf. Happy reading...

http://www.nyc.gov/html/dcp/pdf/bytes/lion_metadata.pdf

[lxbarth]

@colinreilly - thanks. This was very useful.

I also did another review and limited the search to where a Queens and non-Queens style house number coexist for the same street name. This eliminates a lot of false positives in the Rockaways.

The new map is color coded by special address codes "A", "B", or "D" (none others exist).

historicaddress.csv

Addresses where a queens and non-queens house number exists for the same street name on a building. Green: special code A, blue: special code B, red: no special code - code documentation - interactive map

Solution

If we drop A and B addresses as suggested on #116 we also eliminate almost all occurrences of queens and non-queens addresses on the same building. We can go through the rest manually.

[lxbarth]

We're done here after #116 was applied:

If we drop A and B addresses as suggested on #116 we also eliminate almost all occurrences of queens and non-queens addresses on the same building. We can go through the rest manually.

Remaining work on existing data here: #117