<div dir="auto"><div>Scenario one depends on the licensing of the tax data even if it is on paper.<div dir="auto"><br></div><div dir="auto">Remember we live in different countries with different rules.</div><div dir="auto"><br></div><div dir="auto">Have fun but scenario two looks even more doubtful.</div><div dir="auto"><br></div><div dir="auto">Cheerio John</div><br><br><div class="gmail_quote"><div dir="ltr">On Thu, 14 Feb 2019, 3:04 pm Kevin Kenny <<a href="mailto:kevin.b.kenny@gmail.com">kevin.b.kenny@gmail.com</a> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On Thu, Feb 14, 2019 at 10:54 AM John Whelan <<a href="mailto:jwhelan0112@gmail.com" target="_blank" rel="noreferrer">jwhelan0112@gmail.com</a>> wrote:<br>
> By this definition any import of data that has as part of its process each item added that the item is inspected visually using the todo list is not an import.<br>
><br>
> There are a lot of building outlines that are being brought in in this way currently. Are they exempt from the import guidelines?<br>
<br>
There's a whole continuum here, and some room for judgment. Consider<br>
the following:<br>
<br>
1. I'm out mapping in my neighbourhood, and forget to jot down a house<br>
number. When I trace the building footprint, I notice the extra house<br>
and look it up on a paper print of the tax map.<br>
<br>
2. The same scenario, except that I have a database of address points<br>
and look up the address point there.<br>
<br>
3. The same scenario, but I open that database as a separate layer in<br>
JOSM swap layers, and click on the address point.<br>
<br>
4. The same scenario, but I copy-and-paste the house number rather<br>
than retyping it.<br>
<br>
5. The same scenario, but I develop a plugin to search for address<br>
points within the building footprint and offer me the opportunity to<br>
accept or reject them.<br>
<br>
6. The same scenario, but now I do it for all building footprints in a<br>
selected region that don't have addresses already, accepting or<br>
rejecting them one at a time.<br>
<br>
7. The same scenario, only instead of a visible area in JOSM, I use<br>
the entire city or county.<br>
<br>
I think that nearly everyone would agree that 1 is not an import, and<br>
that 7 is. (despite the fact that I'm still offered the opportuinity<br>
to accept or reject addresses.)<br>
<br>
I've been entirely comfortable with up to about item 4. without<br>
discussion, on the grounds that it's not a mechanical edit. Even 5.<br>
doesn't bother me all that much, and I've been tracing enough building<br>
footprints lately that I think it might add value for me to spend a<br>
few hours working that idea up. Beyond that, I'd want to talk, but I'd<br>
expect that given the fact that the data are Public Domain by law,<br>
that applying addresses to existing building footprints that don't<br>
have them would be relatively noncontroversial. (Yes, Frederik, I know<br>
that you'd object!)<br>
<br>
There's room for judgment here. Some would content that if you<br>
consulted any external data source, ever, regarding the feature, that<br>
it's an import. (Some would take the hard line that if I learnt about<br>
the existence of a trail by reading about it in a copyrighted<br>
guidebook, that I'm permanently mentally contaminated and shouldn't<br>
even map that trail in the field!) There are others who would contend<br>
that as long as the data are license-compatible and our handling<br>
comports with the guidelines for mechanical edits, that the<br>
formalities are unnecessary.<br>
<br>
I'm not aware of any successful effort to draw a bright line.<br>
</blockquote></div></div></div>