Hi Paul,<br><br><div class="gmail_quote">On Tue, Dec 11, 2012 at 1:04 PM, Paul Norman <span dir="ltr"><<a href="mailto:penorman@mac.com" target="_blank">penorman@mac.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im"><br></div>Any update strategy has to<br>
be able to handle the case where objects are created in OSM without the IDs<br>
or the IDs get broken in OSM because they can't be verified. Because<br>
anything you do has to handle the case of no IDs, what's the point of them?<br></blockquote><div><br></div><div><div>As a strawman "some houses don't have addresses, therefore the postal service shouldn't use them anywhere" </div>
</div><div><br></div><div>Because it makes it easier? Faster? Less reviews? Easier to query/link data with external systems?</div><div><br></div><div>In New Zealand every road has a SUFI (govt-speak for an ID) - which is used in roading, statistical, electoral, addressing, local govt, and a bunch of other places. Makes sense for OSM to include that key where we can, since it's linkable to other datasets/services.</div>
<div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Lots of people include IDs because they think they might be useful but very<br>
seldom are they actually used.<br></blockquote><div><br></div><div>Agreed - if there's no update strategy and the ID is not enforced/mandated/used/published by the source organisation, then they're definitely pointless.</div>
<div> </div><div>Rob :)</div></div>