<html><head></head><body><div class="ydpef712de9yahoo-style-wrap" style="font-family:times new roman, new york, times, serif;font-size:16px;"><div><div dir="ltr" data-setdir="false">As I see it, one problem with following the RM model of "street:substreet" comes when the data consumer is unaware of the "substreet" concept and ignores it. This leads to 1 Rose Cottages, Green Lane" (addr:housenumber=1; addr:substreet=Rose Cottages; addr:street=Green Lane) being rendered (or used in another way) as 1 Green Lane, when there could well be a genuiine house at 1 Green Lane further along.</div><div dir="ltr" data-setdir="false"><br></div><div dir="ltr" data-setdir="false">So, notjhing is perfect, but I still think that Street / Parentstreet is more useful. </div><div><br></div><div class="ydpef712de9signature"><div style="font-family:new times, serif;font-size:16px;"><div>Regards,</div><div dir="ltr">Peter</div><div dir="ltr"><br></div>(aka PeterPan99)</div></div></div>
<div><br></div><div><br></div>
</div><div id="ydpd907bd79yahoo_quoted_5382012032" class="ydpd907bd79yahoo_quoted">
<div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;">
<div>
On Monday, 14 February 2022, 12:05:29 GMT, Colin Smale <colin.smale@xs4all.nl> wrote:
</div>
<div><br></div>
<div><br></div>
<div><div dir="ltr">> On 02/14/2022 12:04 PM Mark Goodge <<a shape="rect" href="mailto:mark@good-stuff.co.uk" rel="nofollow" target="_blank">mark@good-stuff.co.uk</a>> wrote:<br clear="none">> <br clear="none">> <br clear="none">> On 14/02/2022 09:53, Colin Smale wrote:<br clear="none">> > <br clear="none">> > Rule of thumb if you don't have space in your database or label for two "streets" is to prefer the most specific one and omit the "parent street". The resulting address is sufficiently unique to allow cross-checking with the postcode.<br clear="none">> > <br clear="none">> > That would suggest this possibility:<br clear="none">> > <br clear="none">> > addr:housenumber=2<br clear="none">> > addr:street=Great Moreton Terrace<br clear="none">> > addr:parentstreet=New Road<br clear="none">> > <br clear="none">> > Any software unaware of "addr:parentstreet" would just use "addr:street" which would give a result that is "good enough."<br clear="none">> <br clear="none">> Nominatim won't cope with that, though, as the addr:street value won't <br clear="none">> match the name of a nearby highway.<br clear="none"><br clear="none">That's Nominatim's problem...<br clear="none"><br clear="none">If we start choosing/adjusting our tagging just because of one tool, that's "tagging for the renderer", even if the renderer in question is a core tool in the OSM ecosystem.<br clear="none"> <br clear="none">> Other than that, I agree with you, and what you suggest is actually <br clear="none">> fairly similar to the way that Royal Mail handle it as well as being <br clear="none">> consistent with the more generic "address line 1, address line 2", etc, <br clear="none">> format of many address databases. So long as the first line of the <br clear="none">> address and the postcode are correct, the absence of the rest won't <br clear="none">> significantly affect postal or navigational purposes.<br clear="none"><br clear="none">What would be even closer to the RM model would be using substreet/street instead of street/parentstreet but that idea didn't make the cut.<div class="ydpd907bd79yqt6313945655" id="ydpd907bd79yqtfd49107"><br clear="none"><br clear="none">_______________________________________________<br clear="none">Talk-GB mailing list<br clear="none"><a shape="rect" href="mailto:Talk-GB@openstreetmap.org" rel="nofollow" target="_blank">Talk-GB@openstreetmap.org</a><br clear="none"><a shape="rect" href="https://lists.openstreetmap.org/listinfo/talk-gb" rel="nofollow" target="_blank" fg_scanned="1">https://lists.openstreetmap.org/listinfo/talk-gb</a><br clear="none"></div></div></div>
</div>
</div></body></html>