<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<div class="moz-cite-prefix">On 6/6/2021 10:22 AM, Skyler Hawthorne
via Talk-us-newyork wrote:<br>
</div>
<blockquote type="cite"
cite="mid:63003043-4204-4664-8693-fbf0ee4d2094@dead10ck.com"> <span
dir="ltr" style="font-family:sans-serif">Hey everyone, I'm
pleased to announce that I've finished all counties outside of
NYC! The tally of addresses now stands at: 4,370,743 ...</span>
<br>
<br>
<span dir="ltr" style="font-family:sans-serif">For NYC, it will
require a code update, so it will be a bit before I can generate
the data, but once I get there, since NYC already has its own
previous imports, I'll be scrutinizing the data a bit more
closely to see if there are any special cases to consider with
interactions with the existing NYC data. If it looks fine, then
I'll go ahead and import it. If it's too messy or troublesome,
then I will leave it out now and in the future.</span> <br>
<br>
<span dir="ltr" style="font-family:sans-serif">...</span> <br>
<br>
<span dir="ltr" style="font-family:sans-serif">After NYC is done,
my next steps are to write up a section on the wiki page that
provides some guidelines for cleaning up the data. There have
been a few common questions about how to handle questionable
data, so I think it will help to have this there in the coming
months as people encounter these addresses.</span> <br>
<br>
<span dir="ltr" style="font-family:sans-serif">After that, I'll be
implementing updates on the data. I have so far been doing the
entire import with data from October 2020, and there is already
new data to work with for implementing updates.</span></blockquote>
<p>Congrats Skyler on this impressive import. I was mapping in
Yonkers a few days ago and found myself wishing that the address
import had been completed there to help with some road name
ambiguity. Suddenly the nodes appeared! It was a magic moment.</p>
<p>In your original import proposal you mentioned you might consider
leaving NYC out of the import altogether, since it's already
populated with its own import from city data. If you've changed
your mind about this, I'd ask you to take a very cautious approach
here. I have specific worries if you'd like to hear them, but in
general the issue is that it's super dense with lots of existing
data, and that the import will generate a lot of QA work, and a
lot of that will fall to me. Personally I'd suggest attacking the
QA guidelines next, because I imagine that process will yield some
insight on the kinds of issues that have arisen in denser areas,
which might help in tuning your import code.</p>
<p>Viva NY!<br>
Jason<br>
</p>
</body>
</html>