Hi Koos - I assume you'll be removing the dupe ways from this changeset? Was about to do so, and saw that the change wasn't even 5 mins old.

Hi Quinten - just so you know, you don't want to name the riverbank relation. The waterway name goes onto the waterway=river way. You also don't want to name landuse=residential with the name of the town - that should be done with a point feature.
Hope that makes sense!

Hi there - fyi, you're not using Tertiary and Tertiary Link highways correctly. Tertiary ways should be for arterial routes between villages (or within towns) and should allow for 2-lane travel for 4-wheeled vehicles. Teriary Links connect a Tertiary way to a higher classed route - secondary or primary for instance. Hope that helps...

GR

When ways are merged into reach other - one of the ways will show as "deleted". You can look at the history of way/61885684 (way/61885684/history) and see that all of the nodes actually show they are "part of way Pungwe River (680415942)"

It wasn't *actually* deleted - it was merged into way/680415942

Please don't create fake points of interest. Your additions are being deleted from the map. Please let me know if you have questions - thanks.

Hi Frans - thanks so much for your input. Unfortunately, OSM does not really allow for accurate mapping in the 3rd world - it is aligned solidly with mapping the first world, so I'm trying to map the realities on the ground with the iconography we have.
There is a class of "path" in sub-Saharan Africa where brush has been cleared to allow passage of a 4-wheeled vehicle, but since 4-w vehicles are scarce, the footpath traffic makes it appear that the highway is merely a path. When that path-road is access to and/or within a residential area, I'm trying to find a classification showing that the way resides somewhere in the locus between residential road, path, driveway and living street.
So I'm simply following Russ Deffner's lead: the preferred class for these roads is "service road", due to the way they are usually rendered. You are undoubtedly seeing a lot more service roads as you validate, and you'll be seeing even more in future.
On the second item, I don't actually do that - I connect all my paths going to residential clusters. It drives me nuts when people make highways to building clusters without having them join.
Thanks again for the input!
Greg

You did fine for a first edit! I put barriers at the end of the destroyed roads to make it clear that the road is actually destroyed

Please don't delete and edit features in a task next to yours - you should have noticed that the task to the East of you was checked out!

Hello Andy - looks like you deleted part of the border between Uganda and the DRC in this changeset. The Lamya River *IS* the border along here. Can you fix asap, or back this changeset out?
Many thanks!

Greg

I appreciate the guidance. Last year while HOT mapping in Bangladesh and Nepal, I was directed to start using Living Streets to map alley-like inner roads within populated areas. I had been using service road or alley to denote these - it was a difficult adjustment. Unfortunately now my use of them is somewhat automatic.
I greatly appreciate the feedback and will change my usage accordingly.
Cheers...

btw, for the images you linked to...

The Getty photo is at the Petrapole/Benapole BOP, here: osm.org/search?query=benapole#map=17/23.03615/88.88122. That's a municipal stretch.
The photo from the Star (and yes I saw that article when I was researching) is from this stretch: osm.org/#map=18/26.63535/88.41594. Definitely municipal.
I believe the GQ photo was taken along the Ichamati, near Basirhat. Regardless, it also appears to be municipal, based on the existence of reservoir ponds to the left. I'd bet (although I don't recall reading this) that the border road would also be accessible where it runs alongside a large river like the Ichamati.
Anyway...

Thanks so much for your comments, David. I'm grateful that someone has at least noticed the situation.
At the time I made those edits, there was no fence drawn, just a primary or secondary road marking the entirety of the border fence space. My decision to mark much of that road as access=no was more of a "fail closed" solution to a very complex situation, especially given that the mapping was part of HOT-mandated response to flooding in Bangladesh.
The border road itself was mandated to be used by India's BSF for efficient travel between BOPs (Border Out-Posts), "to facilitate movement of Security Forces", and was largely NOT to be used by locals except in cases where the fence acted as a boundary for a municipality or border haat (bazaar).
(taken from numerous docs from India's Ministry of Home Affairs -
http://mha.nic.in; and the Ministry of Development of Northeast Region - http://www.mdoner.gov.in/node/200)
By 2000, only ~500 miles of border fence had been erected, while almost 2000 miles of roads had been created. So without the fence, those stretches of road were not policed, and became thoroughfares for locals. But as fence construction caught up with the preexisting roads, many (but not all) stretches of road became gated, and only "officially" accessible by BSF.
To complicate matters, I learned on a blog (written by someone helping on the Bangladeshi/Indian border for the British Red Cross) that some stretches of the border road are passable for some distance if you pay a bribe, or if the guards know you, or if you speak a specific dialect of Bangla... or if... blah blah blah.
At that point, frustrated, I said "Fuck it", and stopped researching, and decided to just mark the non-municipal parts as "access=no". The fact that people had repeatedly been shot by the BSF was a major part of that decision - I didn't want a road with dubious access to tempt an aid worker to detour toward possible danger.
At BEST, the border road (in non-municipality, non-BOP and non-haat areas) *might* be open to local foot and bicycle traffic, but barriers do exist to keep out non-BSF motor vehicles. They are especially visible where the fence crosses rivers!
Hopefully that helps clarify things a bit...
I really do think that marking the border roads as Tertiary in this area is not at all reflective of reality - especially when you consider the absolute lack of traffic of ANY kind in newer satellite imagery!
All in all though, I do acknowledge that we can't armchair map this situation. Local mappers NEED to be involved.

Reverting. Apologies, but you can't edit in a locked area!

Many of these objects are fake, and do not exist. User has had over a year to respond to changeset comments - many of the user's object additions have been shown to be bogus (repeating names and derivations of the same name; phone numbers with country/area codes from other places; bus stops on side streets and alleys that couldn't possibly accommodate a bus, etc).
Deleting/reverting.

Objects are fake, and do not exist. User has had over a year to respond to changeset comments - many of the user's object additions have been shown to be bogus (repeating names and derivations of the same name; phone numbers with country/area codes from other places; bus stops on side streets and alleys that couldn't possibly accommodate a bus, etc).
Deleting/reverting.

Objects are fake, and do not exist. User has had over a year to respond to changeset comments - many of the user's object additions have been shown to be bogus (repeating names and derivations of the same name; phone numbers with country/area codes from other places; bus stops on side streets and alleys that couldn't possibly accommodate a bus, etc).
Deleting/reverting.

Objects are fake, and do not exist. User has had over a year to respond to changeset comments - many of the user's object additions have been shown to be bogus (repeating names and derivations of the same name; phone numbers with country/area codes from other places; bus stops on side streets and alleys that couldn't possibly accommodate a bus, etc).
Deleting/reverting.

Objects are fake, and do not exist. User has had over a year to respond to changeset comments - many of the user's object additions have been shown to be bogus (repeating names and derivations of the same name; phone numbers with country/area codes from other places; bus stops on side streets and alleys that couldn't possibly accommodate a bus, etc).
Deleting/reverting.

Objects are fake, and do not exist. User has had over a year to respond to changeset comments - many of the user's object additions have been shown to be bogus (repeating names and derivations of the same name; phone numbers with country/area codes from other places; bus stops on side streets and alleys that couldn't possibly accommodate a bus, etc).
Deleting/reverting.