Talk:UK 2019 Q3 Project: Solar Power

hello - at london city hall we took [photo supply data from https://epc.opendatacommunities.org EPC data] released by ofgem and geocoded it, it is defined as "Percentage of photovoltaic area as a percentage of total roof area. 0% indicates that a Photovoltaic Supply is not present in the property."

Power plant addition

Hi and congrats for this huge project.
Do you find desirable to encourage to add surrounding perimeters, areas, multipolygon or site relations with power=plant for facilities dedicated to electricity production from solar power?
You'll find some examples on the tag page or directly on OSM : 3501886 3501886. Fanfouer (talk) 09:21, 27 August 2019 (UTC)

Hi - thanks. We are indeed going for the solar power=plant as well. We're prioritising smaller installations, for various reasons. But we are indeed doing plants as well - our local consensus has been to map them as perimeter areas. See for example this discussion thread on talk-gb mailing list. --Danstowell (talk) 09:28, 27 August 2019 (UTC)

That's good news, thanks.

No problem with local consensus, but it depends on the situation right? When there are fences, it is possible to map plants as a single closed area or multipolygon relations. Where there are not (but it is still an industrial plant, no point for domestic installations), I don't see any other solution to involve panels in a type=site relation, do you? Fanfouer (talk) 09:50, 27 August 2019 (UTC)

I don't mind strongly. The consensus is loose, and I deal with all these approaches in my data-consumption. In my personal experience the aerial imagery can make it very clear where to draw a polygon, even when it's not clear whether there are fences. --Danstowell (talk) 09:59, 27 August 2019 (UTC)

That's ok for me, there are often fences and this can be checked locally later. Good luck with remaining facilities to map Fanfouer (talk) 10:12, 27 August 2019 (UTC)