Welcome to OpenStreetMap, we are a collaborative volunteer project. Mapping works best when done via a ground survey as there's so much that can be missed on misinterpreted from satellite or aerial imagery. Ground surveyed data is given higher preference, so it's great you're contributing here to improve the data in OpenStreetMap to better reflect the ground truth.

As for why your private driveway is mapped, see osm.wiki/Why_we_won%27t_delete_roads_on_private_property

If it's ever the case that emergency services needed to access your property or those that live further down the road, having these private roads mapped out may help.

Considering this section is on your property, you can also see access=*?uselang=en#List_of_possible_values on other possible access values and what they mean. eg. permissive (access granted, but may be revoked anytime), private (no access permitted), you can also further restrict my mode of transport depending if you restrict all modes include foot, or if you permit walkers but not vehicles, all possible to tag accurately with the right tagging.

(For anyone else interested) the above is from https://www.foresthistory.org.au/Proceedings2004/141.pdf

That's great, and thanks for confirming. This is exactly why I flagged you about this change at changeset/2393286 so you, who originally contributed this way as Belkin Road could be notified about the change to Grimwade Road and given a chance to give feedback about the change. https://osmlab.github.io/osm-deep-history/#/way/40306252

That's my point, in general in OSM we prefer to set the name based on what the signage on the ground says. This whole discussion originated from the fact that the road name was changed based on another map and not from an on the ground survey of the signage, so it was unclear what the ground truth was.

You can't determine this from Google Street View, this is legally unclear and against OSM's policies, you must physically survey or use another compatible source like Mapillary, etc.

why did you delete this? looks like a driveway to me based on the sources you used.

Thanks, any improvements would be great.

> I think it's best to reject this

OSM doesn't have a concept of accept/reject. Changesets are applied immediately and automatically, then if we want to roll back we can either manually restore to the prior state or revert the changeset (either partially or fully).

The only sources we can use are those at osm.wiki/Australian_Data_Sources#Western_Australia which are CC BY 4.0 with a waiver marked as green which includes selected Main Roads WA open data which is CC BY 4.0 and which Main Roads WA have provided OSM a waiver for -> https://catalogue.data.wa.gov.au/dataset/mrwa-road-network. This dataset is already in editors as a background layer, which is fine to use and indicates Grimwade Road, so no problem with that.

travelmap.mainroads.wa.gov.au -> personal non-commercial use only
landgate.wa.gov.au -> all rights reserved
Google Street View -> not permitted

We can't use any of those.

It sounds like neither of us know what signage is on the ground, if any. I've asked the originally contributor of Belkin Road for feedback at changeset/2393286.

Without any other information, or feedback from other mappers it's fine for this one to be renamed per Main Roads WA data, just not from Landgate data.

Regarding Belkin Road at way/40306252/history do you have any feedback on if you still think it should be Belkin Road or as per Main Roads WA which uses "Grimwade Road" ?

There's been some discussion about this at changeset/164793552

There is a Bing Streetside image we can use here which although hard to read seems to say "Coyrecup North" so I'm tempted to therefore set this section as that per Bing Streetside, and per Main Roads WA. It was named 9 years ago so signage may be different now.

We can't use Landgate data, their data is copyright and not compatible with OSM's licensing requirements.

We can use the Main Roads WA data which is CC BY and which we have a waiver for. We even have it as a background imagery layer in JOSM and overlay imagery layer in iD, all out of the box.

In that dataset it's Coyrecup for the southern section only, then Simper from the junction with Coyrecup North Road.

Main Roads WA have released their data under CC BY and have completed the OSMF waiver allowing us to use their data, we even have a background layer available in JOSM and iD out of the box.

In this case Main Roads WA have the name "Grimwade Road" however this doesn't necessarily mean we should go ahead and change it, instead best to check what's signposted on the ground and ask prior mappers for any background.

In OSM it's good practice to map what's on the ground (osm.wiki/Good_practice#Map_what's_on_the_ground) and therefore the road names in OSM may not always match in any particular government dataset, and that's okay and to be expected.

It's been pointed out previously to you at changeset/162379793 that Landgate data can't be used to update data in OSM, their data is copyrighted they don't permit OSM to use their data, as such we can't risk using it.

foot=* is meant to be used for legal accessibility, and legally you can walk on any road except where there is no pedestrian signage like on some motorways.

To indicate there's no infrastructure for pedestrians we can use sidewalk=no (no footpath alongside the road), shoulder=* (is there a shoulder that may provide some space for pedestrians) or even barrier=guard_rail (though not that helpful unless we can somehow tag if there is space to walk outside the guard rail).

Yeah agreed it was wrong before and at least now it's correct in that it's parseable.

I really don't know which is best access=yes or access=no here, so just seeing if you had an opinion.

If I'm making a map to show paths which are open to the public vs those closed/private, I'd show this as open since it's mostly open during the daylight hours when you'd expect to find things open. So I might want access=yes to hint this to data consumers as the default, or we could be agnostic and expect all data consumers to support :conditional and make their own decision on the default.

On the other hand, I think it's also reasonable for the tags to match what's signposted, if it's signposted as "closed 6:30pm to 6am" you might expect the tag value to match as 18:30-06:00, or if it says "open 6am to 6:30pm" you might expect the tag value to match as 06:00-18:30.

This comes up a lot so maybe I could ask for wider community feedback.

but are pedestrians strictly forbidden based on signage or otherwise? or is it just the case of it's not the best idea to walk here?

do you think we should swap the conditional? Some data consumers won't support conditionals, or even if they do they may still default to the non-conditional. In the case of "closed at night", I think it's better to say it's "open, except at night" instead of it's "closed, except during the day".

Thanks. In https://osmlab.github.io/osm-deep-history/#/way/1087666855 AND wasn't what I intended by the ; but I've made changes and think it's correct now.

I've also fixed a few values that used "maxstay" in fee:conditional which should be "stay" per fee:conditional=*

|| vs |yes| is a matter of style, is there a clear consensus? I prefer |yes| since it's a bit more readable and less error prone in getting the wrong number of lanes.

Based on these comments I've reverted to restore the slipway ways, then added access=private based on the tag's introduced with this change.