Talk:Tag:generator:method=photovoltaic

on (unclosed) way=yes ?

https://wiki.openstreetmap.org/w/index.php?title=Tag:generator:method%3Dphotovoltaic&diff=prev&oldid=2754583 saidd it's valid to have PV on a way, exemple ? Marc marc (talk)

A bank of adjacent modules on a solar farm is an example of using this on a way. For example:

https://www.openstreetmap.org/way/682224734#map=17/51.358547/1.087056

That's for a closed way. I can't think of an example where you'd want to use an unclosed way. -- Gregoryw (talk) 06:28, 16 September 2024 (UTC)

A line would be the best for the a tiling array. Currently Key:generator:solar:tracking discourages that, and suggests areas. But that was debatable somewhere.
—— Kovposch (talk) 08:32, 16 September 2024 (UTC)

why a line and not an area ? an array is a area, not a line

I think it's an error/confusion between a non-closed line and an area.

Generator are fine with a node or an area (or a relation for wind turbine site), but it's not a linear feature Marc marc (talk)

We should keep in mind those vertical solar arrays that could be mapped as a line in OSM. I agree most of generators are nodes or areas, but removing affinity with lines may lead to difficulties about edges cases Fanfouer (talk) 12:30, 21 September 2024 (UTC)

vertical installation is a good example indeed! all that's left is to find an example in osm to document this perfectly valid case. Marc marc (talk)

When there's tracking, you can't represent a fixed area. The best case is you are drawing the non-tilted upright position, with the accurate area assuming perfect orthorectification. Most times, you would be incorrectly drawing the smaller tilted area, which will look different when imagery is taken at different times.
man_made=crane has a similar problem, but there's no need to debate doing that here. Arrays are rectangular, and small.
A bridge:movable=* isn't always opening & closing. A tracking array is always changing its tilt. And again, size difference.
To applications, lines mean they don't need to compute a straight skeleton to know the axis of rotation. This can still be inaccurate, when the area size is drawn smaller on tilt. —— Kovposch (talk) 18:03, 21 September 2024 (UTC)

I don't find it interesting to draw a PV on a tracker with a line, this position normally never happens unless you are at the pole Marc marc (talk)