Absolutely! And thinking about it further made me realise that if a road (or a runway!) passes through the course, that’s an important feature, so it should appear on the map. I’d still prefer not to have to pull in huge ways if possible, but I suppose some amount of clipping is inevitable.

As regards the hole moving, I’ve steered away from mapping the pin so far. I think they move often enough that mapping them is nonsensical - “if it’s nailed down, map it” certainly doesn’t apply.

And yes, of course we’re going to have some variation in the way people have mapped courses; that’s just the nature of the beast. My point was aimed more at the proposal for mapping, and improving that before it becomes a widely-adopted standard. The courses I visit, I can improve myself, having walked the holes and photographed features. I’m more than happy to do that work, if the result is better quality data.

Well, there are actually two proposals - “hole as a way” and “hole as an area”. I had thought, for some reason, that “hole as a way” had been superseded by “hole as an area”, but re-reading the discussion I see that there’s just some clarification required as to how the way should be plotted. I’m still left wondering, though, what happens when the hole moves every few days.

My initial problem, though, is there’s no easy way to recursively drill down the relations to get all the nodes and ways required to render any given course - I still end up relying on bounding boxes, which could be problematic if, say, a road crosses a course - I don’t want to pull in the entire road!

Richard, are you still based in Charlbury? Turns out I'm going to be just down the road in Eynsham next weekend (26th-28th). Naturally I'm using OSM data to plot my route. ;)

Indeed. I'm an ardent Free culture advocate anyway and although I was aware of OSM before, it was LUGRadio Live UK that actually gave me the prod to start getting involved. It's just another "open source" project that I'm getting involved with!