Ooops, that should have spotting, not spotted!

Thanks for spotted that that accidentally got moved. I've just corrected. Hopefully it'll re-render soon.

That's certainly the logo that I saw on the sticker at the Pilgrims Way / Spring Lane junction. The reference in the Miners Way link will be several miles SE of here, but it's intriguing that there is / has been at least one other sticker out there. The Jerusalem Way website doesn't seem to show an existing link into the UK yet, given the dashed line, but I wouldn't be surprised if it followed a similar route to the Via Francigena if it does get fully developed.

I've just taken a look, having walked the rectangle on the highways around the playing fields. The only place that I can see Jerusalem Way signed is at the junction between Pilgrims Way and Spring Lane.

This is just the portion that I spotted when I passed several years ago. In the UK I expect it probably follows a similar route to the Via Francigena, but I've not confirmed this on the ground. There may well be a longer route relation that this could be added to, rather having this isolated section.

Next time I'm in the area, I'll try to remember to take a look to see whether there's more extensive signing of the route.

Thanks for pointing that out. I'd misread the sign on the A20 that was actually pointing out that the bridge over Greenway Court Road is 12'3". I've corrected this now. I agree that the M20 and CTRL bridges are both very high.

Ooops. Thanks for pointing out. I've removed them now.

I can only assume that this wasn't so clear in earlier aerial imagery, because I agree that they're clearly not solar panels in the current Bing imagery.

Definitely an ISP. Changed name since. I've updated the name and improved the tagging. I can't see a clear tagging option for this on the wiki, so have gone for office=internet_provider based upon current usage in taginfo

As I recall it, it's a shop front for a local Internet provider that offered Internet access to slightly more remote areas via a microwave link. Although I'm in the UK, as it happens I should be visiting this area in a few weeks' time. I'll try to remember to go and check then -- I've no idea whether it's even still open or located at these premises now, for example.

No worries. I've added the solar panels there now, as you'll have seen. For now it just has the rough geometry and is missing detail such as the module count and orientation, since I was short of time when I added it.

This'll be the journey that it was part of:
http://threecornerscycleride.org.uk/2009/06/27/day-12/
I do have the surveying photos and GPX that I took, which seem to show Gurney Mews (way/37327795) as being a narrow service road with decorative paving stones. So I'd be tempted to adjust the highway=cycleway that's there now to a highway=service with surface=paving_stones (though a little difficult to judge exactly the extent of the surface). At the Commercial Street entrance to Cross Street my survey photo shows a short contraflow cycle lane and a No Motor Vehicles sign (without any exception panel) and some no-entry signs in Cross Street immediately after the junction with Gurney Mews.

I'll email over the GPX and photos that I have over for this stretch to the email address you use on the email lists.

Hi!

First of all thanks for mapping the solar panels here.

I noticed that my solar mapping tool managed to trip up on today's run and it looks like it was because of an oddity with the data in this edit. Specifically the solar panels immediately to the north of North Florida Road, where there were accidentally two overlapping components to a multipolygon. One of the libraries that I'm using in the code behind the site doesn't like the overlaps. So I've just made an edit to fix up the overlap and tweak the way that the multipolygon was made up, in order to hopefully fix the issue for tomorrow morning's run of my site.

The way you'd made up the multipolygon here was by having multiple separate polygons, some sharing common sides with one another, and putting them all in to "outer" roles. This isn't the normal way of doing a multipolygon, so I've adjusted that. The normal way would for this case would have been one single polygon with an "outer" role and then four smaller polygons inside it marked with an "inner" role. I.e. those are "punching holes" in the outer one. I.e. similar to the "One outer and two inner rings" example on this page: osm.wiki/Relation:multipolygon

Here's a link to my update:
changeset/97185302

Since I've only very occasionally used the iD editor, I don't know whether this way of generating the multipolygon may be down to the way that it works.

Anyway, I'm just leaving this comment to let you know that I made the change and why I did it. Looking forward to more solar mapping from you.

Cheers

Gregory

Yes, it's postcode centroids. There's some details here:
osm.wiki/Ordnance_Survey_OpenData#Code-Point_Open
Since these are centroids, rather than the postcode at every individual address, they're not perfect. For example, it's not always clear where the postcode changes along longer roads. However, it's useful for getting coverage of the general bulk of postcodes.

This one was Esri World Imagery, as available in JOSM, as shown in the changeset tags. However, I've been using a mixture of imagery sources for my similar changesets, depending upon which imagery appears clearest or most up-to-date in the area of the country that I'm working in.

Unfortunately there isn't any addressing data with the imagery.

My solar mapping comparison tool will identify the building that a PV installation is over, if we also have the building in OSM. I've also been thinking that we could use the presence of a PV installation mapped with location=roof without an accompanying building to be able to indicate that we still need to map a building there (probably many).

I've been trying to keep the PV installations mapped separately from the buildings for a number of reasons, including:
- They're separate features. Otherwise, for example, it can look like a house itself is a power generator, rather than the modules on it.
- Some buildings have multiple PV installations facing different directions. When combined with module counts or system size data, it's useful to have the greater granularity of detail for estimating the actual output at a given time.

My tool is here:
http://osm.gregorywilliams.me.uk/solar/

Cheers,
Gregory

Ooops, that was a mistake. Sorry. JOSM managed to remember the description from a previous changeset and I managed to commit it with that instead of one describing that I'd added PV installations.

At the time my access to imagery was limited. I've now moved it to a separate way.

The JOSM reverter doesn't seem to be working at the moment, so I've used Frederick Ramm's perl reverter script to perform the revert. A few relations seem to have been edited since (presumably the coastline). I've taken a cursory glance through the boundaries and route relations and they seem to be OK, but I'll try to sanity check the data in more depth after work tonight.

This changeset has been reverted fully or in part by changeset/37014168 where the changeset comment is: Revert deletions made in changeset/36985172 in the Kingsdown area

I think I had marked the relevant squares as closed. I was just adding a note that there were some roads visible in the Bing imagery in adjacent "squares" that could also be added.

I hadn't seen that mailing list post. I'll stop using the Land Registry Price Paid data until it becomes clear that it can be used again. Luckily you'll find that I've tagged all of my direct uses with source:addr:postcode=land_registry_price_paid_data and all of the ones that I've interpolated from existing "knowns" with source:addr:postcode=interpolation.