[Talk-GB] Large swaths of "heath" in Wales?

[Andy Townsend]

On 08/02/2017 20:46, Marco Boeringa wrote:
>
> - Sam888, e.g. http://www.openstreetmap.org/way/413378224
>
> - Glucosamine: http://www.openstreetmap.org/way/405845733
>
> - Dyserth: http://www.openstreetmap.org/way/388818928
>
> There may be more... All of these "users" are prolific, leave almost 
> no changeset comments, and seem to be editing all day. It seems to me 
> these are editors working professionally for some OSM related company.
>

Whilst anything is possible; I'm not convinced.  They don't have that 
"map all of X following a management edict" style that mappers-for-pay 
tend to have.  The sort of things they map differs from each other, and 
at least one of them readily replies to changeset discussion comments:

http://resultmaps.neis-one.org/osm-discussion-comments?uid=2762871
http://resultmaps.neis-one.org/osm-discussion-comments?uid=1302721
http://resultmaps.neis-one.org/osm-discussion-comments?uid=1892838

I'm actually pretty sure that they all actually "mean well" and are 
trying to do the right thing, even if (in the first case at least) their 
edits are mostly misguided.

> Andy: Is there any chance the DWG could figure out which company these 
> people are working for, so the company could be contacted about this 
> specific issue and asked not to add these type of difficult to 
> identify natural features?
>

Ha!  While DWG work sometimes does involve a bit of sleuthing (e.g. 
finding the author of a problem app when there's nothing to go on in the 
changeset info) I think that you may be overestimating our powers here :)


> There are so many changesets involved, I guess doing reverts is almost 
> impossible, lest one wants to see also more useful stuff being removed 
> as well, like roads and large and small patches of forest that I also 
> see being part of these changesets.

I'd tend to agree with that, though the perl revert scripts, which can 
"not touch things since edited by other mappers" and "undo changes made 
by one particular user in particular changesets" might be userful.

>
> I have the feeling the most offending stuff is primarily the false 
> natural=heath. 

... and (in the case of one of those mappers) also tracks.  See the 
discussion on https://www.openstreetmap.org/changeset/33841330 .

> So maybe it is a better course of action to select the heath features 
> in the affected regions in JOSM, and delete only those in a new 
> changeset. I think this is by far the easiest solution. Of course, a 
> bit of caution and review will be required to not include properly 
> digitized heath features by regular OSM users.
>
> Any other ideas?

"nodes last edited by" might be useful too, for example:

http://overpass-turbo.eu/s/lN9

Best Regards,

Andy